Release all 5177 KC classifications, full raw archives, class dictionary, article and 12-slide Wu briefing

.gitattributes

@@ -163,3 +163,23 @@ wu_20261108_active_refinement/manuscript_with_supplement.pdf filter=lfs diff=lfs
 wu_20261108_active_refinement/source_data/external/lateral_voltage_switch_2026.xlsx filter=lfs diff=lfs merge=lfs -text
 wu_20261108_active_refinement/source_data/meshes/BRAIN.glb filter=lfs diff=lfs merge=lfs -text
 wu_20261108_active_refinement/supplementary_information.pdf filter=lfs diff=lfs merge=lfs -text
+wu_20261109_full_atlas_briefing/assets/registered_brain.pdf filter=lfs diff=lfs merge=lfs -text
+wu_20261109_full_atlas_briefing/data/all_class_memberships.csv filter=lfs diff=lfs merge=lfs -text
+wu_20261109_full_atlas_briefing/exports/KC_all_classifications_5177.xlsx filter=lfs diff=lfs merge=lfs -text
+wu_20261109_full_atlas_briefing/exports/manuscript.pdf filter=lfs diff=lfs merge=lfs -text
+wu_20261109_full_atlas_briefing/exports/manuscript_with_supplement.pdf filter=lfs diff=lfs merge=lfs -text
+wu_20261109_full_atlas_briefing/exports/supplementary_information.pdf filter=lfs diff=lfs merge=lfs -text
+wu_20261109_full_atlas_briefing/exports/吴老师汇报_全量KC分类与网络计算_12页_20261109.pdf filter=lfs diff=lfs merge=lfs -text
+wu_20261109_full_atlas_briefing/exports/吴老师汇报_全量KC分类与网络计算_12页_20261109.pptx filter=lfs diff=lfs merge=lfs -text
+wu_20261109_full_atlas_briefing/figures/ExtendedData11.pdf filter=lfs diff=lfs merge=lfs -text
+wu_20261109_full_atlas_briefing/figures/ExtendedData5.pdf filter=lfs diff=lfs merge=lfs -text
+wu_20261109_full_atlas_briefing/figures/ExtendedData6.pdf filter=lfs diff=lfs merge=lfs -text
+wu_20261109_full_atlas_briefing/figures/ExtendedData7.pdf filter=lfs diff=lfs merge=lfs -text
+wu_20261109_full_atlas_briefing/figures/Figure1.pdf filter=lfs diff=lfs merge=lfs -text
+wu_20261109_full_atlas_briefing/figures/journal/ExtendedData5.pdf filter=lfs diff=lfs merge=lfs -text
+wu_20261109_full_atlas_briefing/figures/journal/Figure1.pdf filter=lfs diff=lfs merge=lfs -text
+wu_20261109_full_atlas_briefing/figures/journal/SupplementaryFigure1.pdf filter=lfs diff=lfs merge=lfs -text
+wu_20261109_full_atlas_briefing/figures/journal/SupplementaryFigure2.pdf filter=lfs diff=lfs merge=lfs -text
+wu_20261109_full_atlas_briefing/figures/journal/SupplementaryFigure6.pdf filter=lfs diff=lfs merge=lfs -text
+wu_20261109_full_atlas_briefing/manuscript/main.pdf filter=lfs diff=lfs merge=lfs -text
+wu_20261109_full_atlas_briefing/manuscript/supplementary.pdf filter=lfs diff=lfs merge=lfs -text

wu_20261109_full_atlas_briefing/README.md

@@ -0,0 +1,71 @@
+# 全量KC分类、原始数据与吴老师组会汇报（2026-11-09）
+
+本次交付覆盖全部 **5,177 KC**。这是现有分类版本的统一索引与文章后继版，原始32/15/14、旧候选、失败项、未覆盖项和全部历史发布均保留。
+
+## 师弟从这里开始
+
+1. 打开 `exports/KC_all_classifications_5177.xlsx`。`KC5177` 是全细胞主表；`class_dictionary` 逐类解释；`finest20`、`stable8` 与 `original61` 为常用入口。
+2. 下载 `exports/KC_all_classifications_handoff.zip`，里面有全部 **4,038 个方法版本类别条目**的 CSV/TXT 名单，共 8,076 份名单文件。用 `member_list_index.csv` 对照短文件名 CL00001 与 `namespace::label`。这些条目包含不同方法、K、种子和残余角色，**不是 4,038 种生物类型**。
+3. `data/KC5177_master.csv` / Parquet：每个KC的18位root_id、坐标、原始已知类型、52候选、8稳定、尺度单元、20最细、原61和全部23次全体KC重跑标签；其余内部细分在长表 `all_class_memberships`。
+4. `data/class_dictionary.csv`：每类的大小、来源/尺度、证据状态、主导已知细类×侧别、纯度、覆盖率、坐标质心、前三非KC上下游伙伴、输入家族和MBON输出比例、含义与建议分析。完整方向伙伴计数在 `all_class_partner_fingerprints.parquet`。
+5. `exports/KC_atlas_browser.html` 是不联网的可搜索分类浏览器，可直接复制/下载每类完整ID列表。
+6. 需要真实树、突触坐标或重跑：下载本前缀 `raw/` 两个完整包，或运行 `python code/download_data.py --out downloaded_data`。不要只拿名单当作原始数据。
+
+所有root_id都是**字符串**。Python读取CSV使用 `pd.read_csv(..., dtype={'root_id':str})`，不要经过浮点数；Excel已经写成文本。
+
+## 分类层级
+
+- 12种已有解剖细类：来自 source_hemibrain_type；与原始 official_KC_subtype 一并保留。
+- 52个历史候选构成完整分区。证据状态为 stable_main 965 / coverage_fail 344 / tentative 2,207 / unresolved 1,661 个细胞。
+- 8个跨尺度稳定结构候选覆盖965 KC；9个跨方法共识视图仅作描述，不能再相加算新细胞。
+- 40个选中尺度单元中34确认、6未确认。34的并集2,505；保留全部嵌套、逐尺度判决与失败记录。
+- 20个最细非重叠单元覆盖2,125，另外3,052仍保留。10个纯属同细类×侧别，2个完整恢复KCab-p类型×侧别，最低主导细类纯度67.3%。
+- 全体KC重跑有23次，覆盖4种主表征、K20/40/80/160、初始化与网格敏感性。K80只是预定展示，不是最优功能分类数。
+- 另含3种已知父类型内细分、30次61局部重拟合和历史方法。命名必须带namespace；相同C0在不同方法下没有同一身份。
+
+`namespace_dictionary.csv` 提供每个视图覆盖与重叠计数。原61和focal61只分析该61细胞，其余5,116处于范围之外；不能把它们解释成“没分清的细胞”。算法稳定不等于独立功能验证。
+
+## 完整原始数据与版本
+
+本前缀 raw/：
+
+- `KC_all5177_complete_structural_atlas_20261030.zip`：570,888,101字节，5,301成员，全部5,177个geometry对象，以及KC相关突触、2,261,016端点、有向连接、分支、伙伴注释、分析、代码和审计。
+- `KC_classification_full_sources.zip`：452,747,698字节，32,203原始源文件，含640次多尺度划分、发现/确认标签、协议、完整联合表征、全部拟合、历史对照和原组会纪要。
+
+以上是FlyWire v783 **KC相关子集**，不是把全脑原始电镜发布为本项目数据。每包SHA256见 `metadata/raw_archives.json`，下载脚本校验大小与哈希。包内原始目录结构保留，解压至同一工作目录后即可按原路径访问；代码需pandas/numpy/scipy/pyarrow/scikit-learn/xlsxwriter/openpyxl。
+
+已有全量模拟轨迹和其他公开源对象在 `data/previous_public_data_objects.json` 按不可变HF revision列出（3,271项，约19.7 GB），包含原始/交换时序、全部主/细网格KC与MBON电位、事件、参数、公开生理和跨脑结果。默认下载脚本只取本次两个完整原始包；`--all-historical`再下载历史对象。历史目录中的资料是来源与重放记录，不是本轮新增实验。
+
+## 关键分母与生物对应
+
+- `type_side_purity = 主导已知细类×侧别命中数 / 类成员数`。
+- `dominant_parent_coverage = 命中数 / 该已知细类×侧别全部细胞数`。不是功能准确率。
+- 新指纹的输入分母是全部输入端点，输出分母是全部输出端点，包括KC与未注释伙伴；不同于旧多尺度表的external-only分母。前三伙伴筛掉KC，但给的是计数；完整端点比例仍以全方向为分母。
+- `source_hemibrain_type`细分KCab-s/m/c，不能用粗KCab的分母解释细类型覆盖。
+- 一条ap1原始official标签使用未归一化别名；source细类映射将它保留为ap1。因此新CM020细类表为122/133，旧按归一化official字段为121/132。旧entry表原样保留，差异是口径，不是改了成员。
+- ALPN主要关联嗅觉通路，direct VPN为直接视觉投射，APL为反馈来源，MBON为输出；L为模型操作性间接视觉候选，不能把这些都叫“视觉输入”或用比例直接命名新功能类型。
+
+## 查询与复现
+
+```bash
+python code/query_class.py --namespace finest20 --label SU035 --out SU035_cells.csv
+python code/query_class.py --namespace original61 --label abp_R_K3_C2 --out original14.csv
+python code/query_class.py --root-id 720575940606579465
+python code/build_atlas.py
+python code/build_atlas_figures.py
+python code/build_briefing.py
+python code/update_article.py
+python code/build_paper.py
+```
+
+构建输入见 `metadata/source_manifest.json`；将本项目置于projects/下，运行 `python code/unpack_raw.py /path/to/KC_all5177_complete_structural_atlas_20261030.zip /path/to/KC_classification_full_sources.zip` 映射两个压缩包的目录布局，即可重建。原61与仿真顺序源表已包含在本项目data/中。原始动力学图保留在本目录；其科学源表与重放代码来自不可变前序发布。本文包的独立重放针对分类整合、图谱图和Overleaf编译；没有重新运行历史全部模拟或湿实验。
+
+## 汇报与文章
+
+`exports/吴老师汇报_全量KC分类与网络计算_12页_20261109.pptx`：12页，可编辑文字并嵌SVG，逐页中文讲稿见 `exports/逐页讲稿.md`；PDF直接保留矢量。文章见 `exports/manuscript.pdf`，SI见 `exports/supplementary_information.pdf`，Overleaf包已独立解压编译。
+
+论文主线：真实网络中循环传递把局部启动扩展为群体募集，输入时间组织和树上时序分配进一步改变累积。主/细网格中全部六组交换效应非零，但两组反向，方向/幅度需继续检查；公开生理限定固定参数迁移范围。结构图谱先整理已知对应，混合与未解释部分作为明确问题保留。
+
+## 署名、来源和使用条件
+
+Git作者：ydchen0806 <yindachen@mail.ustc.edu.cn>。未推定论文作者顺序。真实脑/蘑菇体矢量素材来自FlyWire Consortium / Amy Sterling gallery的固定版本，归属与CC BY-NC 4.0来源说明见 `metadata/mesh_source_attribution.json`。第三方数据分别保留其来源使用条件；不对全部组件宣称统一许可证。文章是按Nature研究论文结构组织的当前稿件，发表层级取决于最终证据与评审。

wu_20261109_full_atlas_briefing/SHA256SUMS

@@ -0,0 +1,237 @@
+d520af2fdea624524906510f6714f02b2e34b640bb53fb518ebfe61d75bd8de2  README.md
+da75b27265c0259332e297dfa953d641cf9106980b2567c65d9b12df11191324  STATE.json
+27b07f29a74da38e30a9598f4699e2aeb7bcf7caa38003d16b6d9230f999c40c  assets/figure4_lower.svg
+3417081555e2a949ae8c1b84d3805f4450ca48670c1f1ed421e23427d4f687df  assets/figure4_upper.svg
+bee86627715925517a3177a731286ec3fa32774ee8494af9b4fca9cb45c87320  assets/mushroom_body.pdf
+fd4bb3043180386f09900856b9bc3d44f1adf39894889bc42a7df393cb9afec3  assets/mushroom_body.svg
+729cddbd53c9336bea20494295eabbbaf98ba80ae1caba82547d7d2230e2b159  assets/network_schematic.svg
+22b17adf4cb975856011339dec01fac6864e69d823d868c0cfc83710d08aa337  assets/real_contacts.svg
+939847d95cb0300346eb24c154487428685842f3da1576e5a42c62cabfc41e45  assets/registered_brain.pdf
+915247fb7a1bb5958758b537e9bef54133cec2ad1715dcd368cabf640f13a4dd  assets/registered_brain.svg
+a13d7a59c678170099bac880da20341f0bdefcb928b0631595b38dd114a80a8c  assets/synapse_circuit.pdf
+4ce9c75020d3dbcc9c8b63ebd15c88c854db086168e0d18b09d7c86d34702f36  assets/synapse_circuit.svg
+203ac1f01ba979aa8c8a2f29c51b0228af2aec81196dbff6fe1a761dfdfc9b8f  code/build_atlas.py
+fdf511af313fda60cc9529eda7e24f37e982864f357910120e0f52b2beb733d9  code/build_atlas_figures.py
+7ac8111db444bb957d6426b791527fe6f930c6fa006cfa7dd03c3d557736a1c7  code/build_briefing.py
+f8a45665280fafc162858aa57bda367b89fc544b2338478b8001873fa287710e  code/build_paper.py
+17ec039136a1bd89536ab7fc62766da4c80540e32a26f3fe8f4f0533e7924f6a  code/build_reader_docs.py
+21ff8fa6eb09ce951d1002a8b1586daef1c1ae43f8408d7b5600042c2e2566f3  code/download_data.py
+b7566ac8106cf30f80b0a9b919519b840ab314614394410151d8eb86084db465  code/final_checks.py
+ad7b95e1083dc50ae4019f509a331d38e0573c95358cac6a0b6d3b46e532d9f3  code/journal_layout.py
+70ab8370bb50142948140816462a0cb3fcd76af341aa6b41835bbdd7adb0e6a0  code/package_raw.py
+3aa24075892cc77c0532918144f9c771f341e6c6aa3c7870923d09a8374dbe6c  code/publish_release.py
+e7dc43834d1e669f4cd6a5f2e2952abe1786da5e931428e2af7fceaaccbf820b  code/query_class.py
+7a5f58e2a79984ea23524bfa9e5167626d441549775fc2ec9b72474ab1520745  code/seal_handoff.py
+217608d98c860e43c8031fb4e6016f05873ae2d41a36806fdd50c1b3f404f9b9  code/unpack_raw.py
+590526a826e031f86089eb3abc057a2d97a8b9ed7b94bba4e6f82e7f723be5d6  code/update_article.py
+d5f4ab140edf2b5ed20f6ed77a21a607649828800a01611b85630e8723fd1aa8  data/KC5177_master.csv
+d70331ca8e3cec431d379f96721a673ca706faca43c9d9bd605bb1a16570990e  data/KC5177_master.parquet
+1c5137e776d2a61b990727a9ea1723ad845ff531ac018188e8cbaee668f57a95  data/active_cells.csv
+38ffc12a2d662fddb9abfd796d27364a830b121e6f3ad93946b2b1fcd17f02b4  data/active_pair_summary.csv
+9cfff5bcc92db1f7f16237ef9f0046b708e2ee1f8b7d33459a48a804ae5be7b3  data/all_class_memberships.csv
+90cc3fb3199ba7a1be8c88eddfd8076f55221889fb45d36f3b3aea2a23557ffc  data/all_class_memberships.parquet
+e74a62bddbf9575a2eadca54a167d878177ad4374f54065489340d6aef00cb51  data/all_class_partner_fingerprints.parquet
+f08e141fadbef07cbd14e66c55c57926a971978a3f76c372948b6c26d150015f  data/async_run_summary.csv
+183ee33aa854929d98ec97113dc567ae6962ab84fd1b46033df60c2f49e9629f  data/cell_input_family_profiles.csv
+3ab36189e9cb55ba57877b18ae83868bd45ff8e2e94e441d304aae3472bb4501  data/centroid_runs.csv
+1ea7ecc7743ae858f741ccba451ca788da9cf7bfa478c5c91e20cc82e9f464fa  data/class_dictionary.csv
+80e9ff9fb4834f160b664decd937d02de9e1b21ea778cbbb8a1ebaed76f4896b  data/class_dictionary.parquet
+6532624f99ce6cf118dfa7ffadf73ec8c01dd3ce2013b3811235ad61ce40799d  data/dispersion_runs.csv
+4c4349355151317951be2a15b89624d0cf34187ee7715be129761be700e69c34  data/entry_roles_including_residuals.parquet
+db0d46c4e7dde516306cb21d636848794b9044030bf6fa46dc53903c3deb3f1b  data/finest20_source_summary.csv
+742068cf3d0d39c4c22557f664196ae0974783179cf5a2503854534c31e633b6  data/member_list_index.csv
+ded49af4ab0297de1a80097b6777936573ed1f1725debd476ece01ab41f8d890  data/namespace_dictionary.csv
+d1361e90eff4a5f46d1e1ad7d14c9555765b80a705d68b2dd705c1064d8f4c59  data/numerical_grid_summary.csv
+812f8a2ee492f7d800c1962dd252ce9f5afde5c59b2c9978a5a6bfdd5fda8fa8  data/previous_public_data_objects.json
+2595170a1a914b40c1a134149b3239a6fe1a4098eb7c0a4a25eb39588376099f  data/profile_cells.csv
+1b60821c635703344aa2c23b605949d2f0b3e61d5328f3085e464d56f8ab90c9  data/schedule_masks.csv
+867a87599d1f7c8d0fdb91afa57a02f47e777d94a26f65aa7988cd1876fcd9ec  data/source_original61_with_simulation_order.csv
+bdda097119659aa06f771ed31f824dda2641d71a4e3c2e6b4f94a8f030ebed18  data/stable8_and_consensus9_source.csv
+03f5f3ac1644678174664bfb64c0af510be96d7fe68488fdece905a623c871a8  data/unit_containment.csv
+f342963414d355db774e79842dd3a7f97cc84008839f4ff3fde81499acb4492b  data/unit_scale_decisions.csv
+3b4aa38bbc497fdacd3c8bc97b582a4f32c9b2a73f4c3d51f2ea8be2f2b9b595  exports/KC_all_classifications_5177.xlsx
+690e721be2d6285539aa121d83b34eb90147cf9f347818f223976b1ba6a43a3b  exports/KC_all_classifications_handoff.zip
+f3ea1b1bc2e9e1c372475fbb31a4f48c76968d3a2e5954e2425c6cf5c46c9d23  exports/KC_atlas_browser.html
+ff5fd012ef41141dad95f918af66e3d4d1a5992ef0c50336a55901f777ae682b  exports/briefing_slide_01.png
+5b853d8368e37ab6f7a46839a2ad6f6aee907f55f7180aa343ba49345170496a  exports/briefing_slide_02.png
+872ea463e484e76cfe050b9dcaccd6e0ecb34fb410e0c80f2ad72f5a0b5740d4  exports/briefing_slide_03.png
+38b11538f9982b3e3f9e1fdb790543b2bfbbd481ae04ccc9ec854037680fd0c1  exports/briefing_slide_04.png
+4b531f3d1aba827276b898a9ab8ce61c8f92386fa5f4a61627782d8b67da3f32  exports/briefing_slide_05.png
+6b10765e6459025c83d7638dd0090e08b432fc67ad2d6265a89cd7b62f81c677  exports/briefing_slide_06.png
+4795bb177dbb4d6e976880f78c51b75e266376518967b333ace48fb8ec3869d9  exports/briefing_slide_07.png
+d183f6b4315b97e9bb0a7fac90e2391802af310302b4be9cc2b2548f2da88a01  exports/briefing_slide_08.png
+2f642efc3c3820c9eec55cfe3dc910e6d515a7f14721cc3b5389a7955f06e180  exports/briefing_slide_09.png
+8b9dab0ac2e324e784483f98ec69b0b2927f51b8613a80b065d9a88cf789ad8c  exports/briefing_slide_10.png
+48adf69ac6bc815c4b3a84bf33e417687bb2895d7fb5cebdcf417fe244c2ba33  exports/briefing_slide_11.png
+e5183973730e0f58232cb53420baf833eea768247fdc3569863ae6502f629cda  exports/briefing_slide_12.png
+e867066b3e101f3beaa8e26e84ebf507ed61cd58829330bee0b50316d2e8472e  exports/manuscript.pdf
+0e95f9dfd4099a901782603075a960f611a8714132e8b70b9575f619c700e7fa  exports/manuscript_with_supplement.pdf
+d8f50f240a5a6dbad5e7394915c88e5b750d7231617923ea604bba62e763070f  exports/overleaf_submission.zip
+8cb376bfb31e5b0865ca0f0a29f9ff938938593af785aad2f00aa02e1aa19232  exports/slide_content.json
+b679b5ea94415e9f204e56ff318ecff8c34583a512b56c51b27a706d551f7385  exports/supplementary_information.pdf
+889cb8d19a69b86fe9e338282fdd8fd9d50189f4f8d60a6f999787fe996a1778  exports/吴老师汇报_全量KC分类与网络计算_12页_20261109.pdf
+42e744c3260422d1ef509a7492abd1d094a8f3869fc9c523afddc059b854aeb8  exports/吴老师汇报_全量KC分类与网络计算_12页_20261109.pptx
+cda28d89d88fef716001951f6a0ec848ca3778d4d1d1ebe8861878c39027a5a2  exports/逐页讲稿.md
+b41a92d86a1b6f2d4ba3f34b372bfbbec186c67eaed47a138572d663d7fc7be3  figures/ExtendedData1.pdf
+48f1e48058ae50633e0fd1242d59e9ad7e3cff4550e2a909437194d1f4e166f6  figures/ExtendedData10.pdf
+a53bdfbd02502dee3d463767e755b6138991548bd0eb2791949b66278a475909  figures/ExtendedData11.pdf
+52a910d8271ea39bac8c28f348be04f7da82382ccdccd6a52cde59cd30b600a6  figures/ExtendedData11.svg
+59aa7227ee534f4555d114daee27340bd2ae89ea4799e83964688c8354e40ba2  figures/ExtendedData12.pdf
+1162507c98d51369c0a38d96d6949c56a1d312c5800aeb496872cd6760451f0d  figures/ExtendedData12.svg
+501cf59cf4abc3534a6b4ef2560151e26cb385e76e044d835b21f56267f38fc8  figures/ExtendedData13.pdf
+cc026fd1e85f4966866ad222493a106c168d62c1f84c7075e4a9853d0f834b1a  figures/ExtendedData13.svg
+2aa08f76ed4f2d0e282c74559f1a38e29f9390c8c65876436c0cfb73b6b042e6  figures/ExtendedData14.pdf
+1a318454eac94b7ac07e83653fabb5943ec74888c2a85b8ca25a0f6ae9fefde7  figures/ExtendedData14.svg
+3b24118944979edb6e29f9878cc6d7563f1808f2c8a4aebe302631a20083b35a  figures/ExtendedData15.pdf
+920fd8095fa846b7d20c02b90cd05ceb074f630909c3d17d4e1f506af3112516  figures/ExtendedData15.svg
+7c98314dcab2106996448a4390bf59211faf446a5d344ca333636eb85a5417bc  figures/ExtendedData16.pdf
+717465ff33994a7534a57914d3ad1fdbee8834c5585004e177a91342774bc644  figures/ExtendedData16.svg
+9f1425066b98ab4d5a3d02f6682f6064ee92fa75a6ab53b28a56a23bac9722eb  figures/ExtendedData17.pdf
+1efc7c9558d9cb6fcbe7552caf20c1463157d7bc6b2f6ec897f9494a576d8e99  figures/ExtendedData17.svg
+1609a3724c74c6cd3213536c5fe6feedc1e2ab832ca30bf1eeabf40b24536afb  figures/ExtendedData18.pdf
+145beb621853a2b76d9314cd50d7072e91b6733328d1055eb73ded364536f78f  figures/ExtendedData18.svg
+6de18c14bcd6b83e34e2a11e952aa76277e3a352b0a765d6b99b18c4675e4738  figures/ExtendedData19.pdf
+f33ade03c4ad40d5657870689edb9877e904bc9649f291c376ef0140378cf372  figures/ExtendedData19.svg
+74857b2c59a447d6988507961add9571a4e20aad0f8f3b8d2cacd1ab58019679  figures/ExtendedData2.pdf
+95f05ee457468208914f5c094ce4149ee4712d0d5f0a421888dd01f8731ada3d  figures/ExtendedData20.pdf
+e55360230d422764ed554fd717121722a0c98e6307ee8386ecfa16ec93dc1050  figures/ExtendedData20.svg
+e54013291a456f90a21734ca69aab5181055e6d3341c64607f7d608bb99aa8aa  figures/ExtendedData21.pdf
+7abb92ac3afdcf7975af36aa555d2709f866cc48446972d91bee6a83fec041ef  figures/ExtendedData21.svg
+6c65c69e21341970fb6e57cd6344956341e7adff1bdf54a5ad3d2e61e8d42653  figures/ExtendedData22.pdf
+f99c3b1f256022d3cfd362c1f2b1959fe5cfa38d7696df80448c5ca5b6d605be  figures/ExtendedData22.svg
+40e866721ec122265f75ba4732fd331ee121fb09d23ce3f59bf577fa0d0bcf40  figures/ExtendedData23.pdf
+e3f96471325d710fd993c79a949357e4ac76c306ecb0ba14185f36725a2d5cca  figures/ExtendedData23.svg
+8d4e69eb95e40d9cb22c2ec8c550a836a104d94dd9150b2253501af18cb936f4  figures/ExtendedData24.pdf
+b9ac7e54787f079bcfed202f694709dd103f10800069fab3bfb87644998cb515  figures/ExtendedData24.svg
+76ebcb98df81f6003fb1a583e78705a8dae3bb112ec3e94280988f51fe6ece40  figures/ExtendedData3.pdf
+223eb8b1d6edb00f53ec0ccf7735db7b0f2c5640609dece8396d6f1a0a8e1321  figures/ExtendedData4.pdf
+3e384e198bd6a435a0bd6837d5be386fa7884e8fc43fad932ecf62485f414e2d  figures/ExtendedData5.pdf
+f67a9289dc97c4b93dd7c264cced740630e1d6465d7f7880f5c1d84aebbf833e  figures/ExtendedData6.pdf
+7db429180caddabc694bd7c34c7426b792a062d16a2a81ac5c809a1a2061e021  figures/ExtendedData7.pdf
+c66556e9a5f71cd8e50bedfdf60e2e6d32e445f2d358731abf95e006e1e3d8b6  figures/ExtendedData8.pdf
+b980f10f42ab7826601f2b8b1039709b7c9922cdfe8df4aaff1976fc3624f5db  figures/ExtendedData9.pdf
+84cb3adcf81bb8e303769b8b1d6fdafab3eb318a75ac7b5b55ff69406f86a187  figures/Figure1.pdf
+99ca4523b3bd67c522e2916a43cb5a7db0e08fd234ecc916370526d273149430  figures/Figure1.svg
+83b315fac3d7b7fa78e9035cfc641ac4e49b5e81a3129de7352e42e884598bfe  figures/Figure2.pdf
+3c990b2f1a35442d67686f064a390e86f0f5c85ed21398fa2b69bbbbacc355d3  figures/Figure2.svg
+7eba11e4a68a4a3ce2472fecf7dacc9ffa548e11b3a1decca9e456d6618dfa9b  figures/Figure3.pdf
+8492329b8f22da056df47acc101d984e9af2fc0f563fdeceab31d20b5e452ca6  figures/Figure3.svg
+660c0ae793730cc6d73ca7a20b0f9b21c2a70796666dd87e0ad3969a664b1a7a  figures/Figure4.pdf
+223c34d18f7a38c6f0cb817a51e230f0bdf9be3f41def62796efe7f36af147f7  figures/Figure4.svg
+2a5037f894e36eb12ca33c8d9866e4911221da8967b9f56cb44d5b00f8327baa  figures/Figure5.pdf
+a7add3f23ebdc1dc4e9b104b97178d10cd33814dc2c6b0672b07d6089630b7c8  figures/Figure5.svg
+614379a56a5b5310b1c55efd3e8fc8644f612f18c47b9275e903da7e6d888afe  figures/Figure6.pdf
+f92d1296e73eff75315fe03191ec4ebde6af05ca3f636376d2d6b02cd289fc36  figures/Figure6.svg
+b10d5a0dc192af01a37cdd532cc945d2c47b624e6146ae436d3e4364b9a14739  figures/assignment_mesh_effects.pdf
+76b60c9b211f846afbccb3b9cb56b6b57a15d352b6db9df026703e847972e8ef  figures/assignment_mesh_effects.svg
+5d641a42c8dd5cabfa11bbe9920e5f350510e73b6312b1793919332222cc2e86  figures/atlas_coverage.pdf
+8dc4d14fb6069563f13514ddfcb3c38ffb7ffb5b89cafac6e4ba9fc8524a758a  figures/atlas_coverage.svg
+11f1f87cc9c09a9000f9ca3c1e7418f6e7ab48fe312dca6d7147f806b9d2a2ad  figures/cohort61.pdf
+31d68aad2503fb2cad71d4502470c130f7a38090a7440042275356da101fbf41  figures/cohort61.svg
+eefd796bdb8a299bd40e4d8cab79c955365f69b8010665da57c4c2482cff7bf5  figures/fine20_correspondence.pdf
+282a3fbf26480dd2d2c86d8972dcbaf15bd367a1c53f1f0d1cf76e571183c773  figures/fine20_correspondence.svg
+3d260cf18dd37fa8bf515402adee9371a5ec6e6a5e7cbe564da7972cf2683a5c  figures/fine20_fingerprints.pdf
+b9522fd3f36ea0d89dc2bfde34988c90d5d89ae8129670161a18c4944726fef4  figures/fine20_fingerprints.svg
+b41a92d86a1b6f2d4ba3f34b372bfbbec186c67eaed47a138572d663d7fc7be3  figures/journal/ExtendedData1.pdf
+40e866721ec122265f75ba4732fd331ee121fb09d23ce3f59bf577fa0d0bcf40  figures/journal/ExtendedData10.pdf
+e3f96471325d710fd993c79a949357e4ac76c306ecb0ba14185f36725a2d5cca  figures/journal/ExtendedData10.svg
+74857b2c59a447d6988507961add9571a4e20aad0f8f3b8d2cacd1ab58019679  figures/journal/ExtendedData2.pdf
+76ebcb98df81f6003fb1a583e78705a8dae3bb112ec3e94280988f51fe6ece40  figures/journal/ExtendedData3.pdf
+223eb8b1d6edb00f53ec0ccf7735db7b0f2c5640609dece8396d6f1a0a8e1321  figures/journal/ExtendedData4.pdf
+3e384e198bd6a435a0bd6837d5be386fa7884e8fc43fad932ecf62485f414e2d  figures/journal/ExtendedData5.pdf
+501cf59cf4abc3534a6b4ef2560151e26cb385e76e044d835b21f56267f38fc8  figures/journal/ExtendedData6.pdf
+cc026fd1e85f4966866ad222493a106c168d62c1f84c7075e4a9853d0f834b1a  figures/journal/ExtendedData6.svg
+3b24118944979edb6e29f9878cc6d7563f1808f2c8a4aebe302631a20083b35a  figures/journal/ExtendedData7.pdf
+920fd8095fa846b7d20c02b90cd05ceb074f630909c3d17d4e1f506af3112516  figures/journal/ExtendedData7.svg
+7c98314dcab2106996448a4390bf59211faf446a5d344ca333636eb85a5417bc  figures/journal/ExtendedData8.pdf
+717465ff33994a7534a57914d3ad1fdbee8834c5585004e177a91342774bc644  figures/journal/ExtendedData8.svg
+6c65c69e21341970fb6e57cd6344956341e7adff1bdf54a5ad3d2e61e8d42653  figures/journal/ExtendedData9.pdf
+f99c3b1f256022d3cfd362c1f2b1959fe5cfa38d7696df80448c5ca5b6d605be  figures/journal/ExtendedData9.svg
+84cb3adcf81bb8e303769b8b1d6fdafab3eb318a75ac7b5b55ff69406f86a187  figures/journal/Figure1.pdf
+99ca4523b3bd67c522e2916a43cb5a7db0e08fd234ecc916370526d273149430  figures/journal/Figure1.svg
+83b315fac3d7b7fa78e9035cfc641ac4e49b5e81a3129de7352e42e884598bfe  figures/journal/Figure2.pdf
+3c990b2f1a35442d67686f064a390e86f0f5c85ed21398fa2b69bbbbacc355d3  figures/journal/Figure2.svg
+7eba11e4a68a4a3ce2472fecf7dacc9ffa548e11b3a1decca9e456d6618dfa9b  figures/journal/Figure3.pdf
+8492329b8f22da056df47acc101d984e9af2fc0f563fdeceab31d20b5e452ca6  figures/journal/Figure3.svg
+660c0ae793730cc6d73ca7a20b0f9b21c2a70796666dd87e0ad3969a664b1a7a  figures/journal/Figure4.pdf
+223c34d18f7a38c6f0cb817a51e230f0bdf9be3f41def62796efe7f36af147f7  figures/journal/Figure4.svg
+2a5037f894e36eb12ca33c8d9866e4911221da8967b9f56cb44d5b00f8327baa  figures/journal/Figure5.pdf
+a7add3f23ebdc1dc4e9b104b97178d10cd33814dc2c6b0672b07d6089630b7c8  figures/journal/Figure5.svg
+614379a56a5b5310b1c55efd3e8fc8644f612f18c47b9275e903da7e6d888afe  figures/journal/Figure6.pdf
+f92d1296e73eff75315fe03191ec4ebde6af05ca3f636376d2d6b02cd289fc36  figures/journal/Figure6.svg
+f67a9289dc97c4b93dd7c264cced740630e1d6465d7f7880f5c1d84aebbf833e  figures/journal/SupplementaryFigure1.pdf
+1609a3724c74c6cd3213536c5fe6feedc1e2ab832ca30bf1eeabf40b24536afb  figures/journal/SupplementaryFigure10.pdf
+145beb621853a2b76d9314cd50d7072e91b6733328d1055eb73ded364536f78f  figures/journal/SupplementaryFigure10.svg
+6de18c14bcd6b83e34e2a11e952aa76277e3a352b0a765d6b99b18c4675e4738  figures/journal/SupplementaryFigure11.pdf
+f33ade03c4ad40d5657870689edb9877e904bc9649f291c376ef0140378cf372  figures/journal/SupplementaryFigure11.svg
+95f05ee457468208914f5c094ce4149ee4712d0d5f0a421888dd01f8731ada3d  figures/journal/SupplementaryFigure12.pdf
+e55360230d422764ed554fd717121722a0c98e6307ee8386ecfa16ec93dc1050  figures/journal/SupplementaryFigure12.svg
+e54013291a456f90a21734ca69aab5181055e6d3341c64607f7d608bb99aa8aa  figures/journal/SupplementaryFigure13.pdf
+7abb92ac3afdcf7975af36aa555d2709f866cc48446972d91bee6a83fec041ef  figures/journal/SupplementaryFigure13.svg
+8d4e69eb95e40d9cb22c2ec8c550a836a104d94dd9150b2253501af18cb936f4  figures/journal/SupplementaryFigure14.pdf
+b9ac7e54787f079bcfed202f694709dd103f10800069fab3bfb87644998cb515  figures/journal/SupplementaryFigure14.svg
+7db429180caddabc694bd7c34c7426b792a062d16a2a81ac5c809a1a2061e021  figures/journal/SupplementaryFigure2.pdf
+c66556e9a5f71cd8e50bedfdf60e2e6d32e445f2d358731abf95e006e1e3d8b6  figures/journal/SupplementaryFigure3.pdf
+b980f10f42ab7826601f2b8b1039709b7c9922cdfe8df4aaff1976fc3624f5db  figures/journal/SupplementaryFigure4.pdf
+48f1e48058ae50633e0fd1242d59e9ad7e3cff4550e2a909437194d1f4e166f6  figures/journal/SupplementaryFigure5.pdf
+a53bdfbd02502dee3d463767e755b6138991548bd0eb2791949b66278a475909  figures/journal/SupplementaryFigure6.pdf
+52a910d8271ea39bac8c28f348be04f7da82382ccdccd6a52cde59cd30b600a6  figures/journal/SupplementaryFigure6.svg
+59aa7227ee534f4555d114daee27340bd2ae89ea4799e83964688c8354e40ba2  figures/journal/SupplementaryFigure7.pdf
+1162507c98d51369c0a38d96d6949c56a1d312c5800aeb496872cd6760451f0d  figures/journal/SupplementaryFigure7.svg
+2aa08f76ed4f2d0e282c74559f1a38e29f9390c8c65876436c0cfb73b6b042e6  figures/journal/SupplementaryFigure8.pdf
+1a318454eac94b7ac07e83653fabb5943ec74888c2a85b8ca25a0f6ae9fefde7  figures/journal/SupplementaryFigure8.svg
+9f1425066b98ab4d5a3d02f6682f6064ee92fa75a6ab53b28a56a23bac9722eb  figures/journal/SupplementaryFigure9.pdf
+1efc7c9558d9cb6fcbe7552caf20c1463157d7bc6b2f6ec897f9494a576d8e99  figures/journal/SupplementaryFigure9.svg
+700f527a3d67fbd0d733adf7827589f0b3212c8aa9908cb80a04304c70490bbb  figures/stable_correspondence.pdf
+6c18df0eefab8387fdcfb36dbc5e93a390e018b80b93d8060b65dba23ccc9219  figures/stable_correspondence.svg
+cbf8ae23a93049c40e782c3fc81254846806d8bd2aeaa04d3dd8b8a6c1b70d42  manuscript/article.md
+d5fc80571e4cb80d8dfba1e9f67b2252be94466a67aceb1bdcd251ae94733017  manuscript/legends.json
+e867066b3e101f3beaa8e26e84ebf507ed61cd58829330bee0b50316d2e8472e  manuscript/main.pdf
+e1093e2d6007fec70aa98b1e540cf117d7ac6b2be12c72bcc63b1f5f6c90c62b  manuscript/main.tex
+c62ed68bf353ea9b5b1bbcda0ebeccb1cc82ff4490e6e22073c867e589a1e7c2  manuscript/references.bib
+b679b5ea94415e9f204e56ff318ecff8c34583a512b56c51b27a706d551f7385  manuscript/supplementary.pdf
+12a04a9a685c6b7836e9d13a481164fe61e79923933463477d08d4a9ec9b1da8  manuscript/supplementary.tex
+032de64617a287325b616cbb11f9eac3e2dbb1fd9aa5afd37a7ce71c952dc19c  metadata/atlas_summary.json
+23c94d7e444afd0a6bb4aeecc1aef6152909d34b6cf77c64c9af70ffaf231953  metadata/integration_protocol.json
+3f62faebd25c15c5f93e032d1ad99cbb5abdb9d82c04485fb560009ee377f483  metadata/mesh_source_attribution.json
+e7feee738122e7ea2ce5456fc25f03055b338b0929b63daae267dc247f7a845e  metadata/raw_archives.json
+eeac30c8d89f916edf166a76d8c4386cdc13b4367c43f71e9bb90a1e29b41f1a  metadata/source_manifest.json
+d42fc6c8e82c71e5108d23e2b41f9866a618ecd122f3fe201efd026fb94190af  metadata/verified_references.json
+6bdd47500154b4437fe32c16bcedb7131152d0de55c4e59eca17046768787cb3  validation/article_update_contact_sheet.png
+ba27942107d94d7f658cbfbc0b1ea7983761eae8aad927b3d71e86cdca76bf52  validation/atlas_validation.json
+16da5516e1d4547210693400eb1fe3e709ccacac273825523c4ea907006467af  validation/briefing_validation.json
+b6315e60e0226ca4c1c2470e68682b3f4925b8535b769007419e07027e487c99  validation/browser_logic_check.json
+da9e1d215cc2332a153e535015f479f454b1c969da93827030eac051af81c0d0  validation/extracted_package_compile.txt
+9abd5e4e274d8b5fc5a889434712298f869b6028484092eda3889055dbaad318  validation/final_checks.json
+c30fd7fcbb03d31918ab9f7237b630aa0c1da9197044ca59dd6151034edb6ae1  validation/inherited_raster_inventory.json
+29e40b018765c864c56bf012571fc0d27ca28bd361f343e4b2e47bb0ea0184d8  validation/isolated_atlas_replay.json
+d51c023b3b7e530d6e87ceff3da93c9dcb2a9ee92a8e67b8edfede1b6448a075  validation/manuscript_validation.json
+e20bebf3cb640b5388ea76b4fc59f168deca757ab00665aa2a7b6b5b9a2bb175  validation/new_atlas_paper_page.png
+5515296142b6077297149eebb11b3fe038f1b605c7856e25f92fa5a9aa92e9d6  validation/paper_page_01.png
+3b924768ea17a59f66e21c3d9415cdbd4653ab04f9378d8cac9a7702270b1c04  validation/paper_page_02.png
+91c716a4a50d40989ffb94fa4bf231352b79d4e06b10a4cae5bcd6e2bc8a8900  validation/paper_page_03.png
+d86f305d958c4d00b510b2fcd2783f94aca9f0f79298ce71b628217b3de15d30  validation/paper_page_14.png
+0a9c611e715785780514ac42c7878ec11125ba3dfb56c15b493413414463afde  validation/paper_page_15.png
+6cb147b58efad7da71a7b5454f602a5cbe08fb9e87e6f82f64b7f49204f5c479  validation/paper_page_16.png
+950a8bb897f61a12180be4c40a0d7bc71117ebf8bb973513b249bf451cb09170  validation/paper_page_17.png
+0844797ab835b8fd6d148d989da1a7d20ce3961cfca5a165adca0eb6e4766977  validation/paper_page_18.png
+dac8ee7c5fb77ee0791057d6df1e861b5ff8c5f071613d85787bbcbc8ec49d89  validation/paper_page_19.png
+b1850275a580b5c2316c7632e8934367c85fa0963433a62ea94873c804a1ab5b  validation/paper_page_25.png
+7d014d517ee839009cfc8399f4bb4069034bd3e11c2c31bf7f6f1cae6c297b23  validation/paper_page_26.png
+9441f82fd9546cf6b14bac0b5f0efc0ea9d91fd4d08c3a15e3d5f09454dbd914  validation/paper_page_27.png
+5245a59c3916a8e06b193cfb89ba8ca5a15886731aedac172f2e12c471a34ddf  validation/paper_page_28.png
+574645800fb328d33a99a4cef05a7a9610bfc85199acc340209cad019ea391e6  validation/paper_page_29.png
+34881d92d0f7bf040b99a634e37a0de09e26db651e5630ddecd972b7f581111f  validation/raw_archive_secret_scan.json
+39bb4eb422ecda807a8289920864542074090ce232c336424229a2f31e7a2054  validation/review_manuscript_01.png
+1ab84bb89bb19b1c0df875f604b2cbd447844dbb57a4971402c1d6d6f22641d8  validation/review_manuscript_02.png
+a64d249ce461f90155d69bdbecaa5e4b358f87df391b61dd4e66c225a4ffeb1b  validation/review_manuscript_07.png
+356e80b90f00909293ae911d7a55be7aefcfb220a2bd3e7d1889e1d7360fe003  validation/review_supplementary_information_01.png
+1a8e5501c2bacbe7919e6026608c0b3770366cdedc70478ebca322dfed83bea6  validation/review_supplementary_information_02.png
+5a27dda25afaa1b77f8dbc69477d87abffb82af43d2d71ac82d391b9c3473195  validation/review_supplementary_information_15.png
+fb1a9f968080c9b8f00c6ad33ade6e48f649815ee57e1e4465474d2fc5354659  validation/slide_contact_sheet.png
+a46eefa2ea115102777925fc206e1279b87c43df858fc069738ec31d29f0880f  validation/source_compile.txt
+f03ba92b836bd71fda5d2bd9f7003deb7efe1ddb614612dd9dbbe1c1f3447c2c  validation/supplementary_extracted_compile.txt
+704107943330fd493d824223dd77ef3d33614dc2aed558241b0410466f53fe9f  validation/supplementary_source_compile.txt
+d0a059bef5ee8c0b92ef312bd0438e48a2e95edefcafc554b3e8dc00a26ea8b8  validation/visual_review.json
+cce531bfdb367c6e544f556582036bd7e7b1e1ca5a7ed13509d5eef8a6e2d152  研究方法与结论.md
+a352cb441d141ead220c5f14e344061ff20297683ba40a627d66aa9a2f5bb859  类别说明与分析入口.md

wu_20261109_full_atlas_briefing/STATE.json

@@ -0,0 +1,18 @@
+{
+  "status": "ARTIFACTS_VERIFIED",
+  "scientific_action": "Full atlas integration and class profiles; no reclustering or new animal data",
+  "paper_main_pages": 29,
+  "supplementary_pages": 15,
+  "presentation_slides": 12,
+  "raw_archives_bytes": 1023635799,
+  "classification_entries": 4038,
+  "unique_KCs": 5177,
+  "original32_15_14_preserved": true,
+  "scope": "Current complete KC atlas and classification archive; historical full dynamical data pinned separately",
+  "next_research": [
+    "within-fine-type candidate/residual comparison",
+    "independent connectome identity bridge",
+    "active-condition discretization convergence"
+  ],
+  "submission_status": "Manuscript and briefing prepared; no journal submission performed"
+}

wu_20261109_full_atlas_briefing/assets/registered_brain.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:939847d95cb0300346eb24c154487428685842f3da1576e5a42c62cabfc41e45
+size 192008

wu_20261109_full_atlas_briefing/code/build_atlas.py

@@ -0,0 +1,183 @@
+"""Integrate immutable KC classifications. No clustering or physiological labels.
+Run from repository root with pandas, scipy, pyarrow, xlsxwriter, openpyxl.
+"""
+from pathlib import Path
+import hashlib,json,zipfile,tempfile,shutil,re
+import numpy as np,pandas as pd
+from scipy.sparse import coo_matrix,csr_matrix
+P=Path(__file__).resolve().parents[1];R=P.parents[1]
+A=R/'artifacts/wu_20261030_all5177_structural_atlas'
+M=R/'artifacts/wu_20261026_multiscale_units'
+F=R/'projects/wu_20261102_followup_20261103'
+D=P/'data'
+def sha(p):
+ h=hashlib.sha256()
+ with open(p,'rb') as f:
+  for b in iter(lambda:f.read(8*1024*1024),b''):h.update(b)
+ return h.hexdigest()
+def dump(path,obj):path.write_text(json.dumps(obj,ensure_ascii=False,indent=2)+'\n')
+def main():
+ sources=[]
+ def read(p):
+  sources.append(dict(path=str(p.relative_to(R)),bytes=p.stat().st_size,sha256=sha(p)))
+  return pd.read_parquet(p) if p.suffix=='.parquet' else pd.read_csv(p,dtype={'root_id':str})
+ c=read(A/'registry/all5177_cells.parquet').reset_index(names='root_id');c.root_id=c.root_id.astype(str)
+ c=c.sort_values('root_id').reset_index(drop=True);ids=c.root_id.tolist();lookup={r:i for i,r in enumerate(ids)}
+ assert len(ids)==len(set(ids))==5177
+ c['official_KC_subtype_raw']=c.official_KC_subtype
+ # Preserve original alias in raw column; use fine source annotation for correspondence.
+ classes=[];members=[];namespaces=[]
+ def add(ns,label,roots,status,meaning,scale='',source=''):
+  roots=sorted(set(str(x) for x in roots));assert roots and set(roots)<=set(ids),(ns,label)
+  key=ns+'::'+str(label);assert key not in {v['class_key'] for v in classes}
+  classes.append(dict(class_key=key,namespace=ns,label=str(label),n_cells=len(roots),evidence_status=status,meaning_cn=meaning,scale=str(scale),source=source))
+  members.extend(dict(root_id=r,class_key=key,namespace=ns,label=str(label)) for r in roots)
+ def column(ns,col,status,meaning):
+  for lab,g in c.groupby(col,dropna=False):add(ns,lab,g.root_id,status,meaning,source='all5177_cells.parquet')
+ column('known_fine_anatomy','source_hemibrain_type','ANNOTATION_MAPPING','已有解剖细类；用于对应叶部/层次，不代表本项目新增功能类型。')
+ column('official_annotation_raw','official_KC_subtype_raw','SOURCE_ANNOTATION_RAW','原始官方标注，保留别名与粗粒度，不擅自合并字符串。')
+ column('legacy52','legacy_candidate_id','LEGACY_CANDIDATE','历史52候选完整分区；稳定性状态逐类保留。')
+ column('evidence_state','legacy_evidence_level','EVIDENCE_BOOKKEEPING','历史候选的证据状态分组，不是神经元类别。')
+ for lab,g in c[c.legacy_evidence_level.eq('stable_main')].groupby('legacy_candidate_id'):
+  add('stable8',lab,g.root_id,'CROSS_SCALE_STABLE_CANDIDATE','跨尺度稳定的结构候选；需结合已知解剖细类和侧别解释。')
+ roles=read(A/'registry/entry_roles.parquet');roles.root_id=roles.root_id.astype(str)
+ entries=read(A/'registry/entries.csv');entries.to_csv(D/'stable8_and_consensus9_source.csv',index=False)
+ for _,e in entries[entries.entry_kind.eq('CROSS_METHOD_CONSENSUS')].iterrows():
+  rr=roles[roles.entry_id.eq(e.entry_id)&roles.role.isin(['parent_hit','mixed_member'])]
+  assert len(rr)==e.member_n
+  add('consensus9',e.entry_id,rr.root_id,'DESCRIPTIVE_CROSS_METHOD','跨方法共识视图，含少量混合成员；不是额外独立群体。')
+ roles.to_parquet(D/'entry_roles_including_residuals.parquet',index=False)
+ us=read(M/'registry/units_deduplicated.csv');um=read(M/'registry/unit_members.parquet');um.root_id=um.root_id.astype(str)
+ fine=read(R/'artifacts/wu_20261030_fine_units_integrated_atlas/registry/fine_unit_summary.csv')
+ fine.to_csv(D/'finest20_source_summary.csv',index=False)
+ for _,u in us.iterrows():
+  rr=um[um.unit_id.eq(u.unit_id)];assert len(rr)==u['size']
+  state='CONFIRMED_SCALE_CONDITIONAL' if bool(u.confirmed) else 'NOT_CONFIRMED'
+  add('scale_units40',u.unit_id,rr.root_id,state,'固定尺度下的重初始化和20%删边稳定性；34确认、6未确认；保留嵌套。',u.confirmed_scales)
+  if u.unit_id in set(fine.unit_id):add('finest20',u.unit_id,rr.root_id,state,'确认单元中的最细非重叠展示集；未覆盖细胞仍在全表。',u.confirmed_scales)
+ ud=read(M/'registry/unit_decisions.csv');ud.to_csv(D/'unit_scale_decisions.csv',index=False)
+ read(M/'registry/containment.csv').to_csv(D/'unit_containment.csv',index=False)
+ target=read(F/'results/target61_all_members.csv')
+ for label,g in target.groupby('candidate'):add('original61',label,g.root_id,'FROZEN_ORIGINAL_COHORT','原始右侧KCab-p代表队列；C0=32、C1=15、C2=14；C2解释待定。')
+ # Global fits retain every K, arm, seed and bin-width sensitivity, not only K80.
+ glob=read(F/'results/all5177_assignments.parquet')
+ for (arm,k,seed),g in glob.groupby(['arm','k','seed']):
+  ns=f'global_{arm}_K{k}_s{seed}'
+  for label,q in g.groupby('cluster'):add(ns,f'C{int(label):03}',q.root_id,'EXPLORATORY_REPRESENTATION_PARTITION','位置/形态/伙伴身份/联合落点表征的算法分组；标签仅在本方法、K和种子内有效。',k)
+  assert len(g)==5177 and g.root_id.nunique()==5177
+  c[ns]=c.root_id.map(g.set_index('root_id').cluster).map(lambda x:f'C{int(x):03}')
+ parent=read(F/'parent_refinement/all5177_parent_assignments.parquet')
+ for (arm,typ,side,k,child),g in parent.groupby(['arm','parent_type','side','k','child']):
+  add('parent_refinement_'+arm,f'{typ}|{side}|K{k}|C{child}',g.root_id,'EXPLORATORY_WITHIN_KNOWN_PARENT','已知解剖类型×侧别内部细分；K=ceil(n/25)，未确认为新功能亚型。',k)
+ local=read(F/'parent_refinement/target61_local_assignments.parquet')
+ for (arm,seed,child),g in local.groupby(['arm','seed','child']):
+  add(f'focal61_{arm}_s{seed}',f'C{child}',g.root_id,'EXPLORATORY_FOCAL_REFIT','61细胞内部K3重拟合；不能替换原32/15/14队列。',3)
+ # Earlier methods: retain residuals separately by original scope/parent.
+ historical=list(sorted((R/'artifacts').glob('*/registry/assignments.parquet')))+list(sorted((R/'artifacts/nature_kc_20261102/spatial_explanation_v1').glob('*/assignments.parquet')))
+ for path in historical:
+  h=read(path)
+  if 'root_id' not in h:continue
+  h.root_id=h.root_id.astype(str);assert set(h.root_id)<=set(ids)
+  tag=path.parents[1].name if path.parent.name=='registry' else path.parent.name
+  if 'spatial_explanation' in str(path):tag='spatial_explanation_'+path.parent.name
+  dims=[s for s in ['arm','scope','parent','side','kind','setting','k','seed'] if s in h]
+  labelcol='unit_id' if 'unit_id' in h else ('cluster' if 'cluster' in h else 'label')
+  groups=dims+[labelcol]
+  for key,g in h.groupby(groups,dropna=False):
+   values=dict(zip(groups,key if isinstance(key,tuple) else (key,)));parts=[str(values[t]) for t in dims];raw=str(values[labelcol]);residual=raw in ('','nan','-1')
+   label='|'.join(parts+[('RESIDUAL' if residual else raw)])
+   states=';'.join(sorted(set(g['status' if 'status' in g else 'state'].astype(str)))) if ('status' in g or 'state' in g) else 'DESCRIPTIVE'
+   add('historical_'+tag,label,g.root_id,states,'历史方法对照；残余按原父类/作用域单独保留，不把噪声合并成生物类别。',source=str(path.relative_to(R)))
+ mem=pd.DataFrame(members).drop_duplicates(['root_id','class_key']);cat=pd.DataFrame(classes)
+ assert cat.class_key.is_unique
+ # Unified master retains all memberships, including overlapping and failed sets.
+ for ns in ['stable8','consensus9','scale_units40','finest20','original61']:
+  mm=mem[mem.namespace.eq(ns)].groupby('root_id').label.agg(';'.join);c[ns]=c.root_id.map(mm).fillna('UNASSIGNED')
+ c['confirmed_scale_units']=c.root_id.map(mem[mem.class_key.isin(cat.loc[cat.evidence_status.eq('CONFIRMED_SCALE_CONDITIONAL')&cat.namespace.eq('scale_units40'),'class_key'])].groupby('root_id').label.agg(';'.join)).fillna('UNASSIGNED')
+ c['simulation_gid']=c.root_id.map(pd.read_csv(P/'data/source_original61_with_simulation_order.csv',dtype=str,keep_default_na=False).set_index('root_id')['simulation_gid']).fillna('')
+ # All class profiles use the same full-registry source labels and physical coordinates.
+ i_by_key={k:i for i,k in enumerate(cat.class_key)}
+ rr=mem.class_key.map(i_by_key).to_numpy();cc=mem.root_id.map(lookup).to_numpy()
+ B=coo_matrix((np.ones(len(mem),dtype=np.int64),(rr,cc)),shape=(len(cat),5177)).tocsr()
+ for i,row in cat.iterrows():
+  q=c.iloc[B.indices[B.indptr[i]:B.indptr[i+1]]];counts=q.groupby(['source_hemibrain_type','official_hemisphere']).size().sort_values(ascending=False)
+  typ,side=counts.index[0];n=int(counts.iloc[0]);total=int(((c.source_hemibrain_type==typ)&(c.official_hemisphere==side)).sum())
+  cat.loc[i,'dominant_known_fine_type']=typ;cat.loc[i,'dominant_side']=side;cat.loc[i,'parent_hit_n']=n;cat.loc[i,'parent_total_n']=total
+  cat.loc[i,'type_side_purity']=n/len(q);cat.loc[i,'dominant_parent_coverage']=n/total
+  correspondence='完整恢复已知类型×侧别' if n==total==len(q) else ('已知类型内子集' if n==len(q) else '含混合成员，以主导类型解释')
+  cat.loc[i,'correspondence_cn']=correspondence
+  cat.loc[i,'biological_context_cn']=('α/β叶KC；'+{'KCab-p':'后部(posterior)','KCab-s':'表层(surface)','KCab-m':'中层(middle)','KCab-c':'核心(core)'}.get(typ,'层次见原注释')) if typ.startswith('KCab') else ("α′/β′叶KC；"+('中层(middle)' if typ.endswith('-m') else 'ap分型见原解剖注释') if "KCa'b'" in typ else 'γ叶KC；'+('背侧(dorsal)' if typ=='KCg-d' else '主群或稀有γ亚类，见原注释'))
+  cat.loc[i,'legacy_evidence_state_counts']=json.dumps(q.legacy_evidence_level.value_counts().to_dict(),sort_keys=True)
+  if row.namespace=='legacy52':cat.loc[i,'evidence_status']=';'.join(sorted(q.legacy_evidence_level.unique()))
+  cat.loc[i,'recommended_analysis_cn']='按同侧同细类比较残余；检查伙伴身份、杯区/叶部、形态与落点；功能需独立证据。'
+  for axis in 'xyz':cat.loc[i,f'soma_centroid_{axis}_um']=q[f'soma_{axis}_um'].mean()
+ ep=read(A/'data/endpoints.parquet');ep.root_id=ep.root_id.astype(str);ep.partner_cell_type=ep.partner_cell_type.fillna('UNANNOTATED');ep.partner_cell_class=ep.partner_cell_class.fillna('UNANNOTATED')
+ ep['feature']=ep.direction.astype(str)+'|'+ep.partner_cell_class.astype(str)+'|'+ep.partner_cell_type.astype(str)
+ features=sorted(ep.feature.unique());fmap={x:i for i,x in enumerate(features)}
+ cellcounts=ep.groupby(['root_id','feature'],observed=True).size().reset_index(name='count')
+ E=coo_matrix((cellcounts['count'].to_numpy(),(cellcounts.root_id.map(lookup),cellcounts.feature.map(fmap))),shape=(5177,len(features))).tocsr()
+ Q=(B@E).tocsr();fz=pd.DataFrame([x.split('|',2) for x in features],columns=['direction','partner_class','partner_type'])
+ totals={dr:np.asarray(Q[:,fz.direction.eq(dr).to_numpy()].sum(axis=1)).ravel() for dr in ['in','out']}
+ long=[]
+ for i,row in cat.iterrows():
+  ix=Q.indices[Q.indptr[i]:Q.indptr[i+1]];vv=Q.data[Q.indptr[i]:Q.indptr[i+1]]
+  for fi,n in zip(ix,vv):
+   f=fz.iloc[fi];long.append((row.class_key,f.direction,f.partner_class,f.partner_type,int(n),int(totals[f.direction][i]),float(n/totals[f.direction][i])))
+  for dr in ['in','out']:
+   pick=[(features[fi].split('|',2)[2],int(n)) for fi,n in zip(ix,vv) if fz.iloc[fi].direction==dr and fz.iloc[fi].partner_class!='Kenyon_Cell']
+   pick=sorted(pick,key=lambda x:(-x[1],x[0]))[:3]
+   cat.loc[i,dr+'_top3_nonKC_partner_counts']='; '.join(f'{name}:{n}' for name,n in pick)
+   cat.loc[i,dr+'_endpoint_records']=int(totals[dr][i])
+ pd.DataFrame(long,columns=['class_key','direction','partner_class','partner_type','endpoint_records','denominator_all_direction_endpoints','fraction_of_all_direction_endpoints']).to_parquet(D/'all_class_partner_fingerprints.parquet',index=False)
+ # Compact interpretable family fractions, including OTHER (often recurrent KC).
+ fam=ep[ep.direction.eq('in')].groupby(['root_id','input_family']).size().unstack(fill_value=0).reindex(ids,fill_value=0)
+ out=ep[ep.direction.eq('out')].groupby(['root_id','mbon_output']).size().unstack(fill_value=0).reindex(ids,fill_value=0)
+ for name in fam:
+  nums=np.asarray(B@fam[name].to_numpy()).ravel();cat['input_'+str(name)+'_fraction']=np.divide(nums,totals['in'],out=np.zeros(len(cat)),where=totals['in']!=0)
+ if True in out:
+  nums=np.asarray(B@out[True].to_numpy()).ravel();cat['output_MBON_fraction']=np.divide(nums,totals['out'],out=np.zeros(len(cat)),where=totals['out']!=0)
+ cell_profiles=fam.div(fam.sum(axis=1),axis=0).add_prefix('input_').reset_index();cell_profiles.to_csv(D/'cell_input_family_profiles.csv',index=False)
+ # A per-fit, per-class dictionary is accompanied by scope accounting.
+ for ns,g in mem.groupby('namespace',sort=False):
+  unique=g.root_id.nunique();nr=len(g);ng=g.class_key.nunique()
+  namespaces.append(dict(namespace=ns,n_classes=ng,covered_cells=unique,unassigned_cells=5177-unique,membership_rows=nr,overlap=nr>unique,complete_partition=(nr==unique==5177)))
+ nsdf=pd.DataFrame(namespaces)
+ c.to_csv(D/'KC5177_master.csv',index=False);c.to_parquet(D/'KC5177_master.parquet',index=False)
+ mem.to_csv(D/'all_class_memberships.csv',index=False);mem.to_parquet(D/'all_class_memberships.parquet',index=False)
+ cat.to_csv(D/'class_dictionary.csv',index=False);cat.to_parquet(D/'class_dictionary.parquet',index=False);nsdf.to_csv(D/'namespace_dictionary.csv',index=False)
+ # Text-safe spreadsheet; the long table is supplied separately in CSV/Parquet.
+ with pd.ExcelWriter(P/'exports/KC_all_classifications_5177.xlsx',engine='xlsxwriter',engine_kwargs={'options':{'strings_to_numbers':False,'strings_to_urls':False}}) as writer:
+  for sheet,df in [('KC5177',c),('class_dictionary',cat),('namespaces',nsdf),('finest20',cat[cat.namespace.eq('finest20')]),('stable8',cat[cat.namespace.eq('stable8')]),('original61',target)]:
+   df.to_excel(writer,sheet_name=sheet,index=False);w=writer.sheets[sheet];w.freeze_panes(1,1);w.autofilter(0,0,len(df),len(df.columns)-1);w.set_column(0,len(df.columns)-1,19);w.set_row(0,30)
+   if 'root_id' in df:w.set_column(df.columns.get_loc('root_id'),df.columns.get_loc('root_id'),24,writer.book.add_format({'num_format':'@'}))
+ # Every class gets a CSV + text root list INSIDE the portable archive (short stable filenames).
+ mapping=[]
+ with zipfile.ZipFile(P/'exports/KC_all_classifications_handoff.zip','w',zipfile.ZIP_DEFLATED) as z:
+  for i,row in cat.iterrows():
+   roots=mem.loc[mem.class_key.eq(row.class_key),'root_id'].tolist();code=f'CL{i+1:05}'
+   mapping.append(dict(file_code=code,class_key=row.class_key,n_cells=len(roots)))
+   z.writestr(f'member_lists/{code}.txt','\n'.join(roots)+'\n')
+   z.writestr(f'member_lists/{code}.csv',pd.DataFrame({'root_id':roots}).to_csv(index=False))
+  z.writestr('member_list_index.csv',pd.DataFrame(mapping).to_csv(index=False))
+  for path in sorted(D.glob('*')):z.write(path,'data/'+path.name)
+  z.write(P/'exports/KC_all_classifications_5177.xlsx','KC_all_classifications_5177.xlsx')
+ pd.DataFrame(mapping).to_csv(D/'member_list_index.csv',index=False)
+ # Round-trip and conservation gates.
+ import openpyxl
+ wb=openpyxl.load_workbook(P/'exports/KC_all_classifications_5177.xlsx',read_only=True);ws=wb['KC5177'];rows=ws.iter_rows();header=[v.value for v in next(rows)];ix=header.index('root_id');excel=[r[ix] for r in rows];assert all(x.data_type=='s' for x in excel);assert [x.value for x in excel]==ids
+ fineids=set(mem[mem.namespace.eq('finest20')].root_id);confirmed=set(mem[mem.class_key.isin(cat.loc[cat.namespace.eq('scale_units40')&cat.evidence_status.eq('CONFIRMED_SCALE_CONDITIONAL'),'class_key'])].root_id)
+ assert len(fineids)==2125 and len(confirmed)==2505 and mem[mem.namespace.eq('finest20')].root_id.is_unique
+ assert c.legacy_evidence_level.value_counts().to_dict()==dict(tentative=2207,unresolved=1661,stable_main=965,coverage_fail=344)
+ assert mem[mem.namespace.eq('stable8')].root_id.nunique()==965
+ assert sorted(mem[mem.namespace.eq('original61')].groupby('label').size().tolist())==[14,15,32]
+ with zipfile.ZipFile(P/'exports/KC_all_classifications_handoff.zip') as z:
+  assert z.testzip() is None
+  for m in mapping:
+   got=z.read('member_lists/'+m['file_code']+'.txt').decode().splitlines();want=mem[mem.class_key.eq(m['class_key'])].root_id.tolist();assert got==want
+ # Aggregate descriptive findings, computed directly from the dictionary.
+ fcat=cat[cat.namespace.eq('finest20')];dcat=cat[cat.namespace.eq('global_full_K80_s2026110301')]
+ summary=dict(population=5177,known_fine_anatomical_types=c.source_hemibrain_type.nunique(),legacy_candidates=52,stable_candidates=8,stable_cells=965,consensus_views=9,selected_scale_units=40,confirmed_scale_units=34,unconfirmed_scale_units=6,confirmed_union=len(confirmed),finest_units=20,finest_covered=len(fineids),finest_unassigned=5177-len(fineids),global_fits=23,global_K=[20,40,80,160],namespaces=len(nsdf),dictionary_classes=len(cat),membership_rows=len(mem),fine20_pure_type_side_units=int(fcat.type_side_purity.eq(1).sum()),fine20_exact_type_side_units=int(fcat.correspondence_cn.eq('完整恢复已知类型×侧别').sum()),fine20_min_type_side_purity=float(fcat.type_side_purity.min()),K80_pure_type_side_clusters=int(dcat.type_side_purity.eq(1).sum()),raw_endpoint_records=len(ep),per_class_root_files=len(cat)*2)
+ dump(P/'metadata/atlas_summary.json',summary);dump(P/'metadata/source_manifest.json',sources)
+ dump(P/'validation/atlas_validation.json',dict(status='PASS',**summary,unique_root_IDs=5177,Excel_root_IDs_text=True,all_class_list_roundtrips=True,original_membership_preserved=True,raw_source_hashes_verified=True))
+ print(json.dumps(summary,ensure_ascii=False,indent=2))
+if __name__=='__main__':main()

wu_20261109_full_atlas_briefing/code/build_atlas_figures.py

@@ -0,0 +1,93 @@
+"""Compose publication/presentation vector panels directly as SVG.
+All plotted coordinates/counts come from the released tables. Brain surfaces
+are inherited source-aligned anatomical artwork, with original attribution.
+"""
+from pathlib import Path
+from html import escape
+import pandas as pd,numpy as np,json,cairosvg
+P=Path(__file__).resolve().parents[1];D=P/'data';O=P/'figures'
+C=['#267f83','#b85467','#d29b33','#6f83ae','#8a6b9d','#759a62']
+def tx(x,y,s,size=18,col='#243746',weight='normal',anchor='start'):
+ return f'<text x="{x}" y="{y}" font-family="DejaVu Sans" font-size="{size}" fill="{col}" font-weight="{weight}" text-anchor="{anchor}">{escape(str(s))}</text>'
+def rect(x,y,w,h,col):return f'<rect x="{x}" y="{y}" width="{w}" height="{h}" fill="{col}"/>'
+def line(x1,y1,x2,y2,col='#cad5da',w=1):return f'<path d="M{x1},{y1} L{x2},{y2}" fill="none" stroke="{col}" stroke-width="{w}"/>'
+def dot(x,y,r,col):return f'<circle cx="{x}" cy="{y}" r="{r}" fill="{col}"/>'
+def save(name,body,w=1200,h=490):
+ s=f'<svg xmlns="http://www.w3.org/2000/svg" width="{w}" height="{h}" viewBox="0 0 {w} {h}">'+rect(0,0,w,h,'white')+''.join(body)+'</svg>'
+ (O/(name+'.svg')).write_text(s);cairosvg.svg2pdf(bytestring=s.encode(),write_to=str(O/(name+'.pdf')))
+def main():
+ cat=pd.read_csv(D/'class_dictionary.csv');c=pd.read_csv(D/'KC5177_master.csv',dtype={'root_id':str});fine=cat[cat.namespace.eq('finest20')].copy()
+ # a: classification coverage; b: fine anatomical correspondence; c,d fingerprints
+ b=[tx(20,30,'a  One population, distinct classification questions',23,weight='bold')]
+ rows=[('Known fine anatomy',5177,12,C[0]),('Historical candidate partition',5177,52,'#899da6'),('Cross-scale stable candidates',965,8,C[1]),('Confirmed nested units (union)',2505,34,C[3]),('Finest non-overlapping units',2125,20,C[0])]
+ for i,(name,n,k,col) in enumerate(rows):
+  y=85+i*76;b +=[tx(24,y,name,21),rect(465,y-19,540,24,'#edf1f3'),rect(465,y-19,540*n/5177,24,col),tx(1020,y,f'{n:,} / {k}',19)]
+ b +=[tx(24,470,'Bar: unique cells. Right: cells / classes. Nested memberships are not additive.',16)]
+ save('atlas_coverage',b)
+ b=[tx(20,30,'b  Stable structures map to known anatomical classes',23,weight='bold')]
+ st=cat[cat.namespace.eq('stable8')]
+ for j,(i,r) in enumerate(st.iterrows()):
+  y=83+j*46;b +=[tx(25,y,r.label,21,C[0],weight='bold'),tx(132,y,f'{r.dominant_side[0].upper()}  {r.dominant_known_fine_type}',20),rect(420,y-17,500,21,'#edf1f3'),rect(420,y-17,500*r.dominant_parent_coverage,21,C[0]),tx(940,y,f'{int(r.parent_hit_n)}/{int(r.parent_total_n)}',19),tx(1080,y,f'{r.type_side_purity:.0%}',19)]
+ b +=[tx(420,53,'Known type x side coverage',16),tx(1080,53,'Purity',16),tx(25,473,'CM016 / CM022: partial KCab-s subsets; CM031 / CM033: exact bilateral KCab-p.',17)]
+ save('stable_correspondence',b)
+ b=[tx(20,30,'Twenty finest units: known-type correspondence remains heterogeneous',22,weight='bold')]
+ for j,(_,r) in enumerate(fine.iterrows()):
+  col=j//10;row=j%10;x=25+col*595;y=83+row*39
+  b +=[tx(x,y,r.label,18,C[0],weight='bold'),rect(x+76,y-16,140*r.n_cells/206,17,C[0]),tx(x+228,y,str(r.n_cells),18),tx(x+272,y,f'{r.dominant_side[0].upper()} {r.dominant_known_fine_type}',17),tx(x+496,y,f'{r.type_side_purity:.0%}',17)]
+ b +=[tx(25,476,f'{int(fine.type_side_purity.eq(1).sum())}/20 pure known type x side; 2/20 exact recovery. All remaining memberships are retained.',17)]
+ save('fine20_correspondence',b)
+ b=[tx(20,30,'c  Directed fingerprints separate input families and MBON output',22,weight='bold')]
+ fields=['input_ALPN_fraction','input_APL_fraction','input_VISUAL_PROJECTION_fraction','output_MBON_fraction'];labels=['ALPN / all input','APL / all input','direct VPN / all input','MBON / all output']
+ for k,label in enumerate(labels):b.append(tx(190+k*244,66,label,16))
+ for j,(_,r) in enumerate(fine.iterrows()):
+  y=91+j*17.8;b.append(tx(24,y+12,r.label,13))
+  for k,f in enumerate(fields):
+   v=r[f];t=min(1,v/.55);rgb=tuple(int(245+(base-245)*t) for base in (38,127,131));col='#%02x%02x%02x'%rgb
+   b +=[rect(188+k*244,y,225,15.5,col),tx(302+k*244,y+12,f'{v:.1%}',12,'white' if t>.7 else '#243746',anchor='middle')]
+ b +=[tx(24,477,'All directional endpoints, including KC contacts; partner-specific full counts are released.',16)]
+ save('fine20_fingerprints',b)
+ # Representative cohort: source-aligned coordinates, no inferred function labels.
+ target=pd.read_csv(P.parents[1]/'projects/wu_20261102_followup_20261103/results/target61_all_members.csv',dtype={'root_id':str});m=c.set_index('root_id').loc[target.root_id].copy();m['group']=target.candidate.to_numpy()
+ b=[tx(20,30,'Original 61-cell cohort remains 32 / 15 / 14',23,weight='bold')]
+ xlo,xhi=m.soma_x_um.min()-5,m.soma_x_um.max()+5;ylo,yhi=m.soma_y_um.min()-5,m.soma_y_um.max()+5
+ for i,(g,q) in enumerate(m.groupby('group')):
+  color=C[i];n=len(q)
+  for _,r in q.iterrows():b.append(dot(45+480*(r.soma_x_um-xlo)/(xhi-xlo),382-290*(r.soma_y_um-ylo)/(yhi-ylo),5,color))
+  b +=[dot(670,118+i*76,7,color),tx(690,125+i*76,f'{g.split("_")[-1]}  n = {n}',25,color,weight='bold')]
+ b +=[line(45,382,530,382),line(45,382,45,80),tx(230,420,'Soma x (FAFB14 um)',17),tx(55,66,'Soma y',16),tx(670,360,'All 61: right KCab-p',23,weight='bold'),tx(670,405,'C0: fixed 32-cell network',21),tx(670,444,'C2: preserve the unresolved 14',21,C[1])]
+ save('cohort61',b)
+ # Directly visualize all paired effects, including sign reversals.
+ pair=pd.read_csv(D/'active_pair_summary.csv');b=[tx(20,30,'All six assignment contrasts persist; direction changes in 2/6',23,weight='bold')]
+ b +=[tx(24,69,'Input / timing seed',17),tx(375,69,'Main mesh',18),tx(710,69,'Joint fine mesh',18)]
+ for j,((seed,mask),g) in enumerate(pair.groupby(['seed','mask'])):
+  r=g.set_index('stage');a=int(r.loc['main','count_change']);z=int(r.loc['fine','count_change']);y=121+j*53
+  b +=[tx(30,y,f'{"V" if mask==2 else "L"} / {str(seed)[-2:]}',22),tx(440,y,f'{a:+d}',25,C[3],anchor='middle'),tx(790,y,f'{z:+d}',25,C[0],anchor='middle'),line(490,y-8,728,y-8,C[1] if a*z<0 else '#9caeb5',2)]
+  if a*z<0:b.append(tx(965,y,'reversal',19,C[1]))
+ b +=[tx(24,472,'Paired total waveform-count changes. Fine mesh: 102/192 cell counts change; one 31-to-32 recruitment.',16)]
+ save('assignment_mesh_effects',b)
+ # A dedicated four-panel paper composition keeps text at readable final size.
+ b=[tx(15,30,'a  Coverage of the same 5,177 cells',23,weight='bold')]
+ labels=['Known fine types','Historical candidates','Cross-scale stable','Confirmed unit union','Finest antichain']
+ for i,((_,n,k,col),label) in enumerate(zip(rows,labels)):
+  y=87+i*65;b +=[tx(18,y,label,20),rect(240,y-17,220,22,'#edf1f3'),rect(240,y-17,220*n/5177,22,col),tx(474,y,f'{n:,} / {k}',18)]
+ b +=[tx(18,426,'Unique cells / entries; nested rows overlap.',17)]
+ b +=[tx(625,30,'b  Stable candidates and known classes',23,weight='bold')]
+ for i,(_,r) in enumerate(st.iterrows()):
+  y=85+i*44;b +=[tx(625,y,r.label,18,C[0],weight='bold'),tx(705,y,f'{r.dominant_side[0].upper()} {r.dominant_known_fine_type}',17),tx(900,y,f'{int(r.parent_hit_n)}/{int(r.parent_total_n)}',18),tx(1055,y,f'{r.type_side_purity:.0%}',18)]
+ b +=[tx(900,54,'Coverage',16),tx(1055,54,'Purity',16)]
+ b +=[tx(15,515,'c  Finest units: known fine type and purity',23,weight='bold')]
+ for i,(_,r) in enumerate(fine.iterrows()):
+  x=18+(i//10)*300;y=563+(i%10)*38
+  b +=[tx(x,y,r.label,17,C[0],weight='bold'),tx(x+75,y,f'{r.dominant_side[0].upper()} {r.dominant_known_fine_type}',16),tx(x+223,y,f'{r.type_side_purity:.0%}',17)]
+ b +=[tx(18,971,'10 pure subsets; 2 exact known classes.',17)]
+ b +=[tx(625,515,'d  Full-direction endpoint fractions',23,weight='bold')]
+ for k,l in enumerate(['ALPN','APL','VPN','MBON']):b.append(tx(738+k*110,553,l,17))
+ for i,(_,r) in enumerate(fine.iterrows()):
+  y=574+i*19;b.append(tx(625,y+13,r.label,15))
+  for k,f in enumerate(fields):
+   v=r[f];t=min(1,v/.55);rgb=tuple(int(245+(base-245)*t) for base in (38,127,131));col='#%02x%02x%02x'%rgb
+   b +=[rect(723+k*110,y,100,17,col),tx(773+k*110,y+13,f'{v:.1%}',14,'white' if t>.7 else '#243746',anchor='middle')]
+ b +=[tx(625,986,'Inputs / all input; MBON / all output.',17)]
+ save('ExtendedData24',b,1200,1010)
+ print('Seven new vector compositions generated from canonical class tables.')
+if __name__=='__main__':main()

wu_20261109_full_atlas_briefing/code/build_briefing.py

@@ -0,0 +1,124 @@
+"""Chinese briefing with embedded SVG, editable text/flowchart/native chart.
+
+Modern Office uses SVG images; a PNG fallback supports older viewers. The
+standalone source SVGs permit object-level editing in a vector editor. PDF is
+assembled directly from vector artwork and editable slide text/graphics, not
+from slide screenshots.
+"""
+from pathlib import Path
+import json,tempfile,subprocess,shutil,zipfile
+from io import BytesIO
+import cairosvg,fitz,pandas as pd
+from PIL import Image,ImageDraw
+from pptx import Presentation
+from pptx.util import Inches,Pt
+from pptx.dml.color import RGBColor
+from pptx.enum.shapes import MSO_SHAPE
+from pptx.chart.data import CategoryChartData
+from pptx.enum.chart import XL_CHART_TYPE,XL_TICK_LABEL_POSITION
+from pptx.opc.package import Part
+from pptx.opc.packuri import PackURI
+from pptx.opc.constants import RELATIONSHIP_TYPE as RT
+from pptx.oxml.xmlchemy import OxmlElement
+P=Path(__file__).resolve().parents[1];OUT=P/'exports';NAME='吴老师汇报_全量KC分类与网络计算_12页_20261109'
+INK=(36,55,70);TEAL=(36,127,131);GRAY=(95,112,124);RED=(184,84,103)
+
+def main():
+ prs=Presentation();prs.slide_width=Inches(13.3333);prs.slide_height=Inches(7.5)
+ records=[];svg_records=[];expected=[];texts=[]
+ def journal_text(value):
+  import re
+  from journal_layout import EXTENDED,SUPPLEMENTARY
+  def sub(m):
+   n=int(m[1]);return '扩展图'+str(EXTENDED.index(n)+1) if n in EXTENDED else '补充图'+str(SUPPLEMENTARY.index(n)+1)
+  return re.sub(r'扩展图(\d+)',sub,value)
+ def text(sl,x,y,w,h,value,size=20,color=INK,bold=False):
+  value=journal_text(value)
+  sh=sl.shapes.add_textbox(Inches(x),Inches(y),Inches(w),Inches(h));tf=sh.text_frame;tf.word_wrap=True;tf.margin_top=tf.margin_bottom=0;tf.margin_left=tf.margin_right=0
+  for i,line in enumerate(value.split('\n')):
+   p=tf.paragraphs[0] if i==0 else tf.add_paragraph();p.text=line;p.font.name='WenQuanYi Micro Hei';p.font.size=Pt(size);p.font.color.rgb=RGBColor(*color);p.font.bold=bold;p.space_after=Pt(12)
+  texts[-1].append(value);return sh
+ def slide(title,summary,notes):
+  sl=prs.slides.add_slide(prs.slide_layouts[6]);texts.append([]);text(sl,.55,.3,12.2,.6,title,27,INK,True);text(sl,.58,1.04,12.1,.6,summary,19,TEAL,True);text(sl,.58,7.08,12.1,.24,f'2026-11-09 · 吴老师组会 | 结构图谱 → 网络计算 | {len(prs.slides):02}',10,GRAY)
+  notes=journal_text(notes);sl.notes_slide.notes_text_frame.text=notes;records.append(dict(title=title,summary=summary,notes=notes));return sl
+ def svg(sl,path,x,y,w,h):
+  data=path.read_bytes();png=cairosvg.svg2png(bytestring=data,output_width=1800)
+  # Preserve aspect ratio rather than distorting quantitative axes or geometry.
+  im=Image.open(BytesIO(png));ratio=min(w/im.width,h/im.height);ww=ratio*im.width;hh=ratio*im.height;xx=x+(w-ww)/2;yy=y+(h-hh)/2
+  pic=sl.shapes.add_picture(BytesIO(png),Inches(xx),Inches(yy),width=Inches(ww),height=Inches(hh))
+  part=Part(PackURI(f'/ppt/media/illustration_{len(svg_records)+1}.svg'),'image/svg+xml',sl.part.package,data);rid=sl.part.relate_to(part,RT.IMAGE)
+  blip=pic._element.xpath('.//a:blip')[0];lst=OxmlElement('a:extLst');ex=OxmlElement('a:ext');ex.set('uri','{96DAC541-7B7A-43D3-8B79-37D633B846F1}');node=__import__('lxml.etree',fromlist=['Element']).Element('{http://schemas.microsoft.com/office/drawing/2016/SVG/main}svgBlip',nsmap={'asvg':'http://schemas.microsoft.com/office/drawing/2016/SVG/main'});node.set('{http://schemas.openxmlformats.org/officeDocument/2006/relationships}embed',rid);ex.append(node);lst.append(ex);blip.append(lst)
+  svg_records.append(dict(slide=len(prs.slides),source=str(path.relative_to(P)),x=xx,y=yy,w=ww,h=hh))
+ def bullets(sl,values):text(sl,.68,2.0,4.1,4.8,'\n'.join(values),20)
+
+ # Source-native SVG crops preserve all original panel geometry.
+ from lxml import etree
+ for name,src,box in [('real_contacts','Figure1.svg','0 40 1080 340'),('network_schematic','Figure2.svg','0 40 1080 340')]:
+  root=etree.parse(str(P/'figures'/src)).getroot();root.set('viewBox',box);root.set('width','1080');root.set('height','340');(P/'assets'/(name+'.svg')).write_bytes(etree.tostring(root))
+ sl=slide('从全体KC细分到真实网络的输入整合','分类形成可复用图谱；循环连接与输入时间组织共同决定募集','开场先给结论。吴老师要求：已有解剖对应先整理、统一突触身份与位置、保留树和真实落点、分析32内部共享或互补。全5177个细胞都在本次交付，含未确认与未覆盖项。结构与模型两层证据分别回答问题。')
+ svg(sl,P/'assets/registered_brain.svg',.65,1.9,6.3,4.6)
+ text(sl,7.2,2.05,5.45,4.65,'全量图谱：5,177个KC\n8个稳定候选；34个尺度确认单元\n\n生物对应：20个最细单元\n10个纯属同一细类×侧别，2个完整恢复\n\n网络结论：固定真实结构\n循环扩展响应；时序改变募集与累积',22)
+ sl=slide('分类是多层视图，全体细胞均保留','52个历史候选、34个嵌套确认单元和20个最细单元各有明确用途','图为唯一���胞数，不能把嵌套单元相加。已知细类12种；原官方字段有别名和粗粒度。8稳定965细胞；34确认的并集2505；20最细2125，另3052仍在主表。新版23次全体KC划分覆盖K20/40/80/160，原始标签全部公开。')
+ svg(sl,P/'figures/atlas_coverage.svg',.6,1.9,12.1,4.9)
+ sl=slide('稳定候选首先对应已有解剖类型','CM031/033完整恢复双侧KCab-p；CM016/022是KCab-s内的部分子集','类型映射使用source_hemibrain_type×侧别。CM016为147/310右KCab-s，CM022为114/311左KCab-s；不能拿粗KCab分母799/844解释细类型覆盖。CM012/020含混合成员，展示纯度。CM020因单个原始official别名，新source细类口径为122/133，旧归一化official口径为121/132；原成员不变。所有候选、命中、残余、对侧角色一并交付。')
+ svg(sl,P/'figures/stable_correspondence.svg',.6,1.9,12.1,4.9)
+ sl=slide('20个最细单元显示不同程度的已知类型对应','10/20纯属单一细类×侧别；2/20完整恢复；其余保留混合结构','这页是本轮全量交叉映射的直接结果。20单元41至206细胞，覆盖2125个KC；主导细类×侧别纯度最低67.3%。与旧表粗KCab纯度不同，使用细KCab-s/m/c后可见混合。完整标签和每类解释见Excel的finest20与class_dictionary。')
+ svg(sl,P/'figures/fine20_correspondence.svg',.6,1.9,12.1,4.9)
+ sl=slide('上下游指纹给每个候选具体的分析入口','ALPN、直接视觉投射、APL与MBON分别统计，并公开所有伙伴类型','所有4038个带方法版本的类别条目都计算了伙伴指纹，不是4038种生物类型。该图选20单元。输入分母为该类全部输入端点，输出分母为全部输出端点，均包括KC接触和未知伙伴；不是外部接点分母。每类给前三非KC伙伴及完整计数，不能由这些比例直接命名嗅觉或视觉功能类型。')
+ svg(sl,P/'figures/fine20_fingerprints.svg',.6,1.9,12.1,4.9)
+ sl=slide('原32/15/14保留，14细胞仍是独立分析问题','三组均为右侧KCab-p；不按新版聚类或模拟响应重写原队列','原61来自CM033。C0为32并对应当前仿真gid0到31；C1为15；C2为14。空间散点为原始FAFB14坐标。交付同时保留四种K80标签、父类内细分及30次61局部重拟合，可比较14与15的形态和伙伴指纹；原始分组没有改变。')
+ svg(sl,P/'figures/cohort61.svg',.6,1.9,12.1,4.9)
+ sl=slide('联合表达位置、方向和伙伴身份，检验新增信息','当前约3.06%的表面增益，约99%可由区域或分支条件空模型复现','按吴老师要求已重跑全5177KC。联合特征由输入端点的伙伴身份×方向×空间表达，避免把输出MBON目标作为输入。已知类型×侧别内交叉验证，词表/SVD/尺度仅在训练折拟合。基准误差0.05383898，联合0.05218953；区域/分支条件置换复现98.98%/99.70%表面增益。说明当前关联增益主要由保留边际解释，不能写作功能分类准确率。')
+ svg(sl,P/'assets/real_contacts.svg',.65,1.9,12.0,3.65)
+ text(sl,.85,5.95,11.65,.85,'方法推进：保留真实伙伴—落点关联，以条件空模型检验新增信息，再到真实网络做干预。',23,TEAL,True)
+ sl=slide('关闭循环，群体募集收缩为一个直接驱动细胞','三组接点独立时序：R0仅gid31；R1各输入均可募集32个KC','固定32KC与2真实被动MBON14，1926循环接点、144KC到MBON接点、1356边界接点。十个KC接受平衡外部驱动，22个通过网络参与。A=ALPN，V=直接视觉投射，L=操作性间接视觉候选，不是自然视觉刺激。三个种子A/V/L/ABC循环开关逐项匹配；R0循环电导和电荷严格为零。')
+ svg(sl,P/'assets/network_schematic.svg',.65,1.85,12.0,3.8)
+ text(sl,.9,5.75,11.5,.85,'循环传递把局部启动扩展为群体响应；这与细胞是否拥有输入接点是两个问题。',23,TEAL,True)
+ sl=slide('固定平均到达，压缩时间分散仍阻断募集','三种子×V/L：分散度0与0.5均无接受波形，原分散度1均32/32','每细胞、每输入家族、每时间槽匹配原始质心，保持18事件预算和原始权重。12个缩减分散条件均静默；原分散六组均32响应。0是各细胞/家族局部质心，不是全网统一时刻。该压缩也改变细胞总输入包络；下一页交换实验则严格保留包络。首种子V/L的质心坍缩联合细化仍无响应。')
+ svg(sl,P/'assets/figure4_upper.svg',.6,1.85,12.1,4.9)
+ sl=slide('相同输入包络，接点时序分配仍改变累积','联合细化后6/6总计数差非零；其中2/6与主网格方向相反','所有种子和两输入、原始和交换两臂都联合细化，共12新增主问题模拟和384新细胞读数。每细胞/家族/等权接点内交换完整18事件列，严格保持事件时间—权重多重集、真实树、位置和伙伴。细网格102/192细胞计数改变；L种子10的gid31从0到2接受波形，覆盖31到32。另有6线程重放，192读数完全一致；同一细轨迹采样0.0125改0.025ms有0/384计数变化。两网格显示敏感性��当前方向和幅度不能视为收敛结论。')
+ svg(sl,P/'figures/assignment_mesh_effects.svg',.6,1.9,12.1,4.9)
+ sl=slide('共享响应中保留连续差异；公开生理约束参数迁移','32细胞共同模式约95.5%；固定参数跨Fly-ID仍高于经验波形基线','二值共享不等于轨迹相同。395/496细胞对的方向距离大于3倍观测细化差异，这是数值分离度，不是生物显著性。公开24ABF、480sweeps，7Fly-ID留出；固定模型RMSE35.216mV，经验基线32.723mV；作者QC后15.820对14.192。记录没有与本KCab-p亚型对齐，因此用作参数迁移约束。本页图为主图6上半部的真实动物留出与比较。')
+ bullets(sl,['同一二值支持集内\n仍有连续轨迹差异','395/496对超过\n3倍观测细化差异','7个Fly-ID留出\n35.22 对 32.72 mV','公开生理限定\n当前参数迁移范围'])
+ svg(sl,P/'figures/Figure6.svg',5.0,1.9,7.75,4.9)
+ sl=slide('把可对应的结构与可干预的计算结果形成文章','数据交给师弟逐类分析；论文围绕真实网络中输入协调与募集展开','汇报结束强调已完成结果与接续分析。全量发布含5177主表、4038方法版本类别说明、8076份逐类名单文件、完整结构与分类原始包，原始动力学给固定HF版本入口。4038是含不同K/种子/方法和残余的条目数。下一步优先同侧同细类匹配的候选—残余比较、另一连接组身份桥接、活跃条件再细化并检查轨迹/计数/方向收敛。遵循吴老师已有对应先写、未解释保留的要求。')
+ rows=[('结构结论','已知类型能对应；细分单元的纯度与覆盖各异'),('计算结论','循环扩展募集；输入时间组织改变响应与累积'),('师弟交付','5,177主表＋全部分类名单＋类别解释＋原始数据'),('文章与下一步','六张主图；补充全分类图谱；做类型内对照与活跃网格收敛')]
+ for i,(a,b) in enumerate(rows):
+  y=2.05+i*1.0;text(sl,.78,y,2.1,.6,a,23,TEAL,True);text(sl,3.0,y,9.5,.76,b,23);shape=sl.shapes.add_shape(MSO_SHAPE.RECTANGLE,Inches(.75),Inches(y+.83),Inches(11.9),Inches(.012));shape.fill.solid();shape.fill.fore_color.rgb=RGBColor(222,230,234);shape.line.fill.background()
+ assert len(prs.slides)==12
+ for slide_obj in prs.slides:
+  for sh in slide_obj.shapes:
+   try:sh.shadow.inherit=False
+   except (AttributeError,NotImplementedError):pass
+   for ref in sh._element.xpath('.//a:effectRef'):ref.set('idx','0')
+ ppt=OUT/(NAME+'.pptx');prs.save(ppt)
+ # Render Office for compatibility; retain vector reassembly as the delivery PDF.
+ with tempfile.TemporaryDirectory(prefix='wu_vector_slides_') as td:
+  td=Path(td);shutil.copy2(ppt,td/'briefing.pptx');res=subprocess.run(['libreoffice','-env:UserInstallation='+(td/'profile').as_uri(),'--headless','--convert-to','pdf','--outdir',str(td),str(td/'briefing.pptx')],capture_output=True,text=True,timeout=90)
+  assert res.returncode==0 and (td/'briefing.pdf').exists(),res.stdout+res.stderr
+  raw=fitz.open(td/'briefing.pdf');assert len(raw)==len(prs.slides)
+  out=fitz.open()
+  for i,pg in enumerate(raw):
+   out.insert_pdf(raw,from_page=i,to_page=i);target=out[-1]
+   # Replace raster fallback regions with direct vector SVG PDFs while retaining
+   # native Office text/flowcharts/chart around them.
+   for row in [v for v in svg_records if v['slide']==i+1]:
+    box=fitz.Rect(row['x']*72,row['y']*72,(row['x']+row['w'])*72,(row['y']+row['h'])*72)
+    target.add_redact_annot(box,fill=(1,1,1));target.apply_redactions(images=1,graphics=2);b=cairosvg.svg2pdf(url=str(P/row['source']));vector=fitz.open(stream=b,filetype='pdf');target.show_pdf_page(box,vector,0)
+   expected.append(''.join(pg.get_text().split()))
+  pdf=OUT/(NAME+'.pdf');out.save(pdf,garbage=4,deflate=True);out.close();raw.close()
+  (P/'validation/office_render.log').write_text(res.stdout+res.stderr)
+ checks=[]
+ with fitz.open(pdf) as doc:
+  sheet=Image.new('RGB',(1440,430*((len(doc)+1)//2)),'#e9edf0');draw=ImageDraw.Draw(sheet)
+  for i,page in enumerate(doc):
+   actual=''.join(page.get_text().split()).replace('−','-')
+   missing=[line for value in texts[i] for line in value.split('\n') if ''.join(line.split()).replace('−','-') not in actual]
+   assert not missing,(i+1,missing)
+   image=page.get_pixmap(matrix=fitz.Matrix(1,1));image.save(OUT/f'briefing_slide_{i+1:02}.png');im=Image.open(OUT/f'briefing_slide_{i+1:02}.png').convert('RGB');im.thumbnail((700,394));sheet.paste(im,((i%2)*720+10,(i//2)*430+5));draw.text(((i%2)*720+10,(i//2)*430+405),str(i+1),fill='black');checks.append(dict(slide=i+1,all_textbox_lines_present=True,raster_images=len(page.get_image_info())))
+  sheet.save(P/'validation/slide_contact_sheet.png')
+ with zipfile.ZipFile(ppt) as z:
+  embedded=[n for n in z.namelist() if n.endswith('.svg')];assert len(embedded)==len(svg_records)
+ report=dict(status='PASS',slides=len(prs.slides),embedded_SVG=len(embedded),native_editable_charts=0,native_editable_workflow=True,actual_office_render=True,delivery_PDF='Office text/shapes plus original vector figure recomposition',slide_checks=checks)
+ (P/'validation/briefing_validation.json').write_text(json.dumps(report,indent=2)+'\n');(OUT/'slide_content.json').write_text(json.dumps(records,ensure_ascii=False,indent=2)+'\n');print(json.dumps(report))
+if __name__=='__main__':main()

wu_20261109_full_atlas_briefing/code/build_paper.py

@@ -0,0 +1,160 @@
+"""Build a reader PDF and a clean, independently compiling Overleaf package."""
+from pathlib import Path
+import re,json,subprocess,shutil,tempfile,zipfile,unicodedata,hashlib
+import fitz
+from journal_layout import records
+P=Path(__file__).resolve().parents[1];R=P.parents[1]
+DOIS=['10.7554/eLife.62576','10.1038/s41467-024-49616-z','10.1038/s41586-024-07763-9','10.1038/s41586-024-07939-3','10.1038/s41586-024-07686-5','10.1016/j.cell.2026.08.015','10.1113/JP288790','10.7554/eLife.56954','10.7554/eLife.77578','10.1038/nn.3547','10.3389/fphys.2026.1898182','10.1016/j.cub.2026.01.014']
+EXT=[('wu_20261105_crossbrain_transfer','crossbrain_transfer'),('wu_20261104_combination_network','anatomical_coverage'),('wu_20261104_combination_network','full_primary_refinement'),('wu_20261105_drive_recurrence','continuous_profiles'),('wu_20261104_public_validation','all21_heldout_traces'),('wu_20261104_public_validation','raw_WT_QC'),('wu_20261104_public_validation','raw_NaChBac_QC'),('wu_20261104_public_validation','native_variants'),('wu_20261104_public_validation','observation_and_calibration'),('wu_20261104_public_validation','published_perturbation_source')]
+
+def esc(s):
+ trans={'±':r'$\pm$','γ':'gamma','μ':r'$\mu$','µ':r'$\mu$','Ω':r'$\Omega$','−':'-','–':'--','—':'---','×':r'$\times$','≤':r'$\leq$','≥':r'$\geq$','→':r'$\rightarrow$','↓':r'$\downarrow$','’':"'",'‘':"'",'“':'"','”':'"','⁻':'-','⁶':'6','α':r'$\alpha$','Δ':r'$\Delta$','²':'2','³':'3','π':r'$\pi$','…':r'$\ldots$'}
+ out=''
+ for c in s:
+  if c in trans:out+=trans[c]
+  else:out+= {'\\':r'\textbackslash{}','&':r'\&','%':r'\%','$':r'\$','#':r'\#','_':r'\_','{':r'\{','}':r'\}','~':r'\textasciitilde{}','^':r'\textasciicircum{}'}.get(c,c)
+ return out
+
+def inline(s):
+ parts=re.split(r'(\[\d+(?:,\d+)*\]|`[^`]+`|https://[^\s]+|\d+(?:\.\d+)? × 10−\d+)',s);out=''
+ for t in parts:
+  if re.fullmatch(r'\[\d+(?:,\d+)*\]',t):out+=r'\citep{'+','.join('ref'+x for x in t[1:-1].split(','))+'}'
+  elif t.startswith('`'):out+=esc(t[1:-1])
+  elif t.startswith('https://'):out+=r'\url{'+t+'}'
+  elif (scientific:=re.fullmatch(r'(\d+(?:\.\d+)?) × 10−(\d+)',t)):
+   out+='$'+scientific[1]+r'\times10^{-'+scientific[2]+'}$'
+  else:out+=esc(t)
+ return out
+
+def compile_at(path,label,basename="main",bibliography=True):
+ logs=[]
+ commands=[['pdflatex','-interaction=nonstopmode','-halt-on-error',basename+'.tex']]
+ if bibliography:commands.append(['bibtex',basename])
+ commands += [['pdflatex','-interaction=nonstopmode','-halt-on-error',basename+'.tex']]*2
+ for cmd in commands:
+  r=subprocess.run(cmd,cwd=path,capture_output=True,text=True,timeout=90);logs.append(r.stdout+r.stderr)
+  if r.returncode:raise RuntimeError(' '.join(cmd)+'\n'+logs[-1][-3500:])
+ (P/f'validation/{label}_compile.txt').write_text('\n'.join(logs))
+ return (path/(basename+'.pdf')).read_bytes(),(path/(basename+'.log')).read_text(errors='replace')
+
+def main():
+ ref_file=R/'projects/wu_20261104_manuscript_release/metadata/verified_references.json'
+ if not ref_file.exists():ref_file=P/'metadata/verified_references.json'
+ refs={**json.loads(ref_file.read_text()),**json.loads((P/'metadata/verified_references.json').read_text())};bib=[]
+ for i,doi in enumerate(DOIS,1):
+  v=refs[doi];author=' and '.join(unicodedata.normalize('NFKD',x).encode('ascii','ignore').decode() for x in v['authors'][:3])+(' and others' if len(v['authors'])>3 else '')
+  title=unicodedata.normalize('NFKD',v['title']).encode('ascii','ignore').decode()
+  bib.append('@article{ref'+str(i)+',\n author={'+author+'},\n title={{'+title+'}},\n journal={'+v['journal']+'},\n year={'+v['year']+'},\n doi={'+doi+'}\n}')
+ (P/'manuscript/references.bib').write_text('\n\n'.join(bib)+'\n')
+ (P/'metadata/verified_references.json').write_text(json.dumps({doi:refs[doi] for doi in DOIS},indent=2)+'\n')
+ for i,(project,name) in enumerate(EXT,1):
+  source_pdf=R/f'projects/{project}/figures/{name}.pdf';target_pdf=P/f'figures/ExtendedData{i}.pdf'
+  if source_pdf.exists():shutil.copy2(source_pdf,target_pdf)
+  assert target_pdf.exists(),target_pdf
+ source=(P/'manuscript/article.md').read_text();title=source.splitlines()[0][2:];abstract=source.split('## Abstract\n\n')[1].split('\n\n## Introduction')[0]
+ text=source.split('## Introduction\n\n',1)[1];body=[]
+ for line in text.splitlines():
+  if line.startswith('## '):
+   name=line[3:]
+   if name=='Methods':body.append('\\clearpage\n\\onecolumn\n')
+   body.append(r'\section*{'+esc(name)+'}')
+  elif line.startswith('### '):body.append(r'\subsection*{'+esc(line[4:])+'}')
+  elif line:body.append(inline(line)+'\n')
+  else:body.append('\n')
+ header=r'''\documentclass[10pt,a4paper]{article}
+\usepackage[T1]{fontenc}
+\usepackage[utf8]{inputenc}
+\usepackage{lmodern}
+\usepackage[margin=15mm,columnsep=7mm]{geometry}
+\usepackage{graphicx,microtype,amsmath}
+\usepackage[super,sort&compress]{natbib}
+\usepackage{xurl}
+\usepackage[colorlinks=true,allcolors=black]{hyperref}
+\usepackage{caption}
+\captionsetup{font=small,labelfont=bf}
+\setlength{\parindent}{1em}
+\setlength{\parskip}{0.25em}
+\setlength{\emergencystretch}{1em}
+\makeatletter
+\renewcommand\section{\@startsection{section}{1}{0pt}{2ex plus 1ex}{1ex}{\normalfont\large\bfseries}}
+\renewcommand\subsection{\@startsection{subsection}{2}{0pt}{1.8ex plus .8ex}{.7ex}{\normalfont\normalsize\bfseries}}
+\makeatother
+\begin{document}
+\twocolumn[{
+\begin{center}{\LARGE\bfseries '''+esc(title)+r'''}\par\vspace{3mm}
+{\small 9 November 2026}
+\end{center}
+\noindent\textbf{'''+inline(abstract)+r'''}\par\vspace{6mm}
+}]
+'''
+ tex=header+''.join(body)+r'\clearpage\bibliographystyle{unsrtnat}\bibliography{references}'+'\n'
+ legends=json.loads((P/'manuscript/legends.json').read_text())
+ layout=records();alias=P/'figures/journal';alias.mkdir(exist_ok=True)
+ for row in layout:
+  for suffix in ['.pdf','.svg']:
+   original=P/'figures'/(row['source']+suffix)
+   if original.exists():shutil.copy2(original,alias/(row['name']+suffix))
+ def figure_tex(row):
+  return ('\\clearpage\n\\noindent\\textbf{'+row['label']+'}\n\\begin{center}\n'
+      +'\\includegraphics[width=\\textwidth,height=.70\\textheight,keepaspectratio]{../figures/journal/'+row['name']+'.pdf}\n'
+      +'\\end{center}\n\\noindent{\\small '+inline(legends[row['source']])+'}\n')
+ for row in layout:
+  if row['section']!='supplementary':tex+=figure_tex(row)
+ supplementary=header.split('\\begin{document}',1)[0]+'\\begin{document}\n\\onecolumn\n'
+ supplementary+='\\begin{center}{\\LARGE\\bfseries Supplementary Information}\\par\\vspace{4mm}\n'+esc(title)+'\\end{center}\n'
+ supplementary+='\\noindent Supplementary Figures 1--14 preserve complete recording, assay, timing and numerical controls. All original model cells and historical outcomes remain.\n'
+ for row in layout:
+  if row['section']=='supplementary':supplementary+=figure_tex(row)
+ supplementary+='\\end{document}\n'
+ (P/'manuscript/supplementary.tex').write_text(supplementary)
+ tex+='\\end{document}\n';(P/'manuscript/main.tex').write_text(tex)
+ pdf,log=compile_at(P/'manuscript','source');(P/'exports/manuscript.pdf').write_bytes(pdf)
+ supp,supplog=compile_at(P/'manuscript','supplementary_source','supplementary',False);(P/'exports/supplementary_information.pdf').write_bytes(supp)
+ assert not any(re.search(r'Overfull|undefined|Float too large|LaTeX Warning',line) for line in supplog.splitlines())
+ with fitz.open(stream=pdf,filetype='pdf') as combined,fitz.open(stream=supp,filetype='pdf') as sd:
+  combined.insert_pdf(sd);combined.save(P/'exports/manuscript_with_supplement.pdf',garbage=4,deflate=True)
+ warnings=[line for line in log.splitlines() if re.search(r'Overfull|undefined|Float too large|LaTeX Warning',line)]
+ assert not warnings,warnings
+ stage=P/'exports/overleaf';
+ if stage.exists():shutil.rmtree(stage)
+ (stage/'manuscript').mkdir(parents=True);(stage/'figures/journal').mkdir(parents=True)
+ for f in ['main.tex','supplementary.tex','references.bib']:shutil.copy2(P/'manuscript'/f,stage/'manuscript'/f)
+ for f in sorted((P/'figures/journal').glob('*.pdf')):shutil.copy2(f,stage/'figures/journal'/f.name)
+ (stage/'README.txt').write_text('Set manuscript/main.tex as the root document and compile with pdfLaTeX/BibTeX. Set manuscript/supplementary.tex as root for the independent supplementary document (pdfLaTeX). Original SVG artwork and numerical source tables are distributed in the full project release.\n')
+ zipfile_path=P/'exports/overleaf_submission.zip'
+ with zipfile.ZipFile(zipfile_path,'w',zipfile.ZIP_DEFLATED) as z:
+  for f in sorted(stage.rglob('*')):
+   if f.is_file():z.write(f,f.relative_to(stage))
+ with tempfile.TemporaryDirectory(prefix='wu_paper_replay_') as tmp:
+  with zipfile.ZipFile(zipfile_path) as z:z.extractall(tmp)
+  replay,replaylog=compile_at(Path(tmp)/'manuscript','extracted_package')
+  assert not any(re.search(r'Overfull|undefined|Float too large|LaTeX Warning',l) for l in replaylog.splitlines())
+  replay_supp,replay_supp_log=compile_at(Path(tmp)/'manuscript','supplementary_extracted','supplementary',False)
+  assert not any(re.search(r'Overfull|undefined|Float too large|LaTeX Warning',line) for line in replay_supp_log.splitlines())
+  with fitz.open(stream=replay_supp,filetype='pdf') as a,fitz.open(stream=supp,filetype='pdf') as b:
+   assert len(a)==len(b) and [page.get_text() for page in a]==[page.get_text() for page in b]
+  with fitz.open(stream=replay,filetype='pdf') as a,fitz.open(stream=pdf,filetype='pdf') as b:
+   assert len(a)==len(b)
+   assert [p.get_text() for p in a]==[p.get_text() for p in b]
+ with fitz.open(stream=pdf,filetype='pdf') as doc:
+  full=' '.join(page.get_text() for page in doc)
+  for marker in ['5,177','559','35.216','32.723','15.820','14.192','95.4852']:assert marker in full,marker
+  assert not re.search(r'/unify/|/tmp/|/root/',full)
+  for stale in (P/'validation').glob('paper_page_*.png'):stale.unlink()
+  for i in sorted(set([0,1,2]+[j for j,p in enumerate(doc) if p.get_text().startswith('Figure ') or any(p.get_text().startswith('Extended Data Figure '+str(n)) for n in [6,7,8,9,10])])):doc[i].get_pixmap(matrix=fitz.Matrix(1.25,1.25)).save(P/f'validation/paper_page_{i+1:02}.png')
+  report=dict(status='PASS',pages=len(doc),main_figures=6,extended_figures=10,supplementary_figures=14,supplementary_pages=len(fitz.open(stream=supp,filetype='pdf')),combined_pages=len(fitz.open(stream=pdf,filetype='pdf'))+len(fitz.open(stream=supp,filetype='pdf')),main_text_words=len(source.split('## Introduction\n\n')[1].split('## Methods')[0].split()),abstract_words=len(abstract.split()),separate_bibliography=True,verified_references=12,overleaf_extracted_compile=True,overleaf_text_identical=True,supplementary_extracted_compile=True,supplementary_text_identical=True,latex_warnings=warnings,visible_local_paths=False,authors_not_inferred=True,main_figure_raster_images=sum(len(fitz.open(P/f'figures/Figure{i}.pdf')[0].get_images()) for i in range(1,7)))
+ with fitz.open(stream=pdf,filetype='pdf') as doc:
+  inherited=[]
+  for i,page in enumerate(doc):
+   images=page.get_image_info()
+   if images:
+    label=page.get_text().splitlines()[0];assert label in ['Extended Data Figure 2','Extended Data Figure 4'],label
+    inherited.append(dict(page=i+1,label=label,images=[dict(width=im['width'],height=im['height']) for im in images]))
+  inventory=dict(status='DOCUMENTED',main_figure_rasters=0,new_extended_rasters=0,inherited_extended_objects=inherited)
+  (P/'validation/inherited_raster_inventory.json').write_text(json.dumps(inventory,indent=2)+'\n')
+ with fitz.open(stream=supp,filetype='pdf') as sd:
+  assert not re.search(r'/unify/|/tmp/|/root/',' '.join(page.get_text() for page in sd))
+  assert all(not page.get_image_info() for page in sd)
+ report['all_supplementary_pages_vector']=True
+ (P/'validation/manuscript_validation.json').write_text(json.dumps(report,indent=2)+'\n');print(json.dumps(report))
+if __name__=='__main__':main()

wu_20261109_full_atlas_briefing/code/build_reader_docs.py

@@ -0,0 +1,103 @@
+"""Write the collaborator entry points from the canonical tables."""
+from pathlib import Path
+import json,pandas as pd,zipfile
+P=Path(__file__).resolve().parents[1]
+def table(df,cols):
+ rows=['| '+' | '.join(cols)+' |','| '+' | '.join(['---']*len(cols))+' |']
+ for _,r in df.iterrows():rows.append('| '+' | '.join(str(r[k]).replace('|',' / ') for k in cols)+' |')
+ return '\n'.join(rows)
+def main():
+ c=pd.read_csv(P/'data/class_dictionary.csv');s=json.loads((P/'metadata/atlas_summary.json').read_text());m=pd.read_parquet(P/'data/all_class_memberships.parquet');ns=pd.read_csv(P/'data/namespace_dictionary.csv')
+ ns['scope_note_cn']=ns.namespace.map(lambda x:'仅原61细胞；其余细胞在该分析范围之外' if (x=='original61' or x.startswith('focal61_') or 'spatial_explanation' in x) else '见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能')
+ ns.to_csv(P/'data/namespace_dictionary.csv',index=False)
+ readme=f'''# 全量KC分类、原始数据与吴老师组会汇报（2026-11-09）
+
+本次交付覆盖全部 **5,177 KC**。这是现有分类版本的统一索引与文章后继版，原始32/15/14、旧候选、失败项、未覆盖项和全部历史发布均保留。
+
+## 师弟从这里开始
+
+1. 打开 `exports/KC_all_classifications_5177.xlsx`。`KC5177` 是全细胞主表；`class_dictionary` 逐类解释；`finest20`、`stable8` 与 `original61` 为常用入口。
+2. 下载 `exports/KC_all_classifications_handoff.zip`，里面有全部 **{s['dictionary_classes']:,} 个方法版本类别条目**的 CSV/TXT 名单，共 {s['per_class_root_files']:,} 份名单文件。用 `member_list_index.csv` 对照短文件名 CL00001 与 `namespace::label`。这些条目包含不同方法、K、种子和残余角色，**不是 {s['dictionary_classes']:,} 种生物类型**。
+3. `data/KC5177_master.csv` / Parquet：每个KC的18位root_id、坐标、原始已知类型、52候选、8稳定、尺度单元、20最细、原61和全部23次全体KC重跑标签；其余内部细分在长表 `all_class_memberships`。
+4. `data/class_dictionary.csv`：每类的大小、来源/尺度、证据状态、主导已知细类×侧别、纯度、覆盖率、坐标质心、前三非KC上下游伙伴、输入家族和MBON输出比例、含义与建议分析。完整方向伙伴计数在 `all_class_partner_fingerprints.parquet`。
+5. `exports/KC_atlas_browser.html` 是不联网的可搜索分类浏览器，可直接复制/下载每类完整ID列表。
+6. 需要真实树、突触坐标或重跑：下载本前缀 `raw/` 两个完整包，或运行 `python code/download_data.py --out downloaded_data`。不要只拿名单当作原始数据。
+
+所有root_id都是**字符串**。Python读取CSV使用 `pd.read_csv(..., dtype={{'root_id':str}})`，不要经过浮点数；Excel已经写成文本。
+
+## 分类层级
+
+- 12种已有解剖细类：来自 source_hemibrain_type；与原始 official_KC_subtype 一并保留。
+- 52个历史候选构成完整分区。证据状态为 stable_main 965 / coverage_fail 344 / tentative 2,207 / unresolved 1,661 个细胞。
+- 8个跨尺度稳定结构候选覆盖965 KC；9个跨方法共识视图仅作描述，不能再相加算新细胞。
+- 40个选中尺度单元中34确认、6未确认。34的并集2,505；保留全部嵌套、逐尺度判决与失败记录。
+- 20个最细非重叠单元覆盖2,125，另外3,052仍保留。10个纯属同细类×侧别，2个完整恢复KCab-p类型×侧别，最低主导细类纯度67.3%。
+- 全体KC重跑有23次，覆盖4种主表征、K20/40/80/160、初始化与网格敏感性。K80只是预定展示，不是最优功能分类数。
+- 另含3种已知父类型内细分、30次61局部重拟合和历史方法。命名必须带namespace；相同C0在不同方法下没有同一身份。
+
+`namespace_dictionary.csv` 提供每个视图覆盖与重叠计数。原61和focal61只分析该61细胞，其余5,116处于范围之外；不能把它们解释成“没分清的细胞”。算法稳定不等于独立功能验证。
+
+## 完整原始数据与版本
+
+本前缀 raw/：
+
+- `KC_all5177_complete_structural_atlas_20261030.zip`：570,888,101字节，5,301成员，全部5,177个geometry对象，以及KC相关突触、2,261,016端点、有向连接、分支、伙伴注释、分析、代码和审计。
+- `KC_classification_full_sources.zip`：452,747,698字节，32,203原始源文件，含640次多尺度划分、发现/确认标签、协议、完整联合表征、全部拟合、历史对照和原组会纪要。
+
+以上是FlyWire v783 **KC相关子集**，不是把全脑原始电镜发布为本项目数据。每包SHA256见 `metadata/raw_archives.json`，下载脚本校验大小与哈希。包内原始目录结构保留，解压至同一工作目录后即可按原路径访问；代码需pandas/numpy/scipy/pyarrow/scikit-learn/xlsxwriter/openpyxl。
+
+已有全量模拟轨迹和其他公开源对象在 `data/previous_public_data_objects.json` 按不可变HF revision列出（3,271项，约19.7 GB），包含原始/交换时序、全部主/细网格KC与MBON电位、事件、参数、公开生理和跨脑结果。默认下载脚本只取本次两个完整原始包；`--all-historical`再下载历史对象。历史目录中的资料是来源与重放记录，不是本轮新增实验。
+
+## 关键分母与生物对应
+
+- `type_side_purity = 主导已知细类×侧别命中数 / 类成员数`。
+- `dominant_parent_coverage = 命中数 / 该已知细类×侧别全部细胞数`。不是功能准确率。
+- 新指纹的输入分母是全部输入端点，输出分母是全部输出端点，包括KC与未注释伙伴；不同于旧多尺度表的external-only分母。前三伙伴筛掉KC，但给的是计数；完整端点比例仍以全方向为分母。
+- `source_hemibrain_type`细分KCab-s/m/c，不能用粗KCab的分母解释细类型覆盖。
+- 一条ap1原始official标签使用未归一化别名；source细类映射将它保留为ap1。因此新CM020细类表为122/133，旧按归一化official字段为121/132。旧entry表原样保留，差异是口径，不是改了成员。
+- ALPN主要关联嗅觉通路，direct VPN为直接视觉投射，APL为反馈来源，MBON为输出；L为模型操作性间接视觉候选，不能把这些都叫“视觉输入”或用比例直接命名新功能类型。
+
+## 查询与复现
+
+```bash
+python code/query_class.py --namespace finest20 --label SU035 --out SU035_cells.csv
+python code/query_class.py --namespace original61 --label abp_R_K3_C2 --out original14.csv
+python code/query_class.py --root-id 720575940606579465
+python code/build_atlas.py
+python code/build_atlas_figures.py
+python code/build_briefing.py
+python code/update_article.py
+python code/build_paper.py
+```
+
+构建输入见 `metadata/source_manifest.json`；将本项目置于projects/下，运行 `python code/unpack_raw.py /path/to/KC_all5177_complete_structural_atlas_20261030.zip /path/to/KC_classification_full_sources.zip` 映射两个压缩包的目录布局，即可重建。原61与仿真顺序源表已包含在本项目data/中。原始动力学图保留在本目录；其科学源表与重放代码来自不可变前序发布。本文包的独立重放针对分类整合、图谱图和Overleaf编译；没有重新运行历史全部模拟或湿实验。
+
+## 汇报与文章
+
+`exports/吴老师汇报_全量KC分类与网络计算_12页_20261109.pptx`：12页，可编辑文字并嵌SVG，逐页中文讲稿见 `exports/逐页讲稿.md`；PDF直接保留矢量。文章见 `exports/manuscript.pdf`，SI见 `exports/supplementary_information.pdf`，Overleaf包已独立解压编译。
+
+论文主线：真实网络中循环传递把局部启动扩展为群体募集，输入时间组织和树上时序分配进一步改变累积。主/细网格中全部六组交换效应非零，但两组反向，方向/幅度需继续检查；公开生理限定固定参数迁移范围。结构图谱先整理已知对应，混合与未解释部分作为明确问题保留。
+
+## 署名、来源和使用条件
+
+Git作者：ydchen0806 <yindachen@mail.ustc.edu.cn>。未推定论文作者顺序。真实脑/蘑菇体矢量素材来自FlyWire Consortium / Amy Sterling gallery的固定版本，归属与CC BY-NC 4.0来源说明见 `metadata/mesh_source_attribution.json`。第三方数据分别保留其来源使用条件；不对全部组件宣称统一许可证。文章是按Nature研究论文结构组织的当前稿件，发表层级取决于最终证据与评审。
+'''
+ (P/'README.md').write_text(readme)
+ fine=c[c.namespace.eq('finest20')][['label','n_cells','dominant_known_fine_type','dominant_side','type_side_purity','dominant_parent_coverage','correspondence_cn','in_top3_nonKC_partner_counts','out_top3_nonKC_partner_counts']].copy()
+ doc='# 每类是什么意思：生物对应与分析入口\n\n每个方法版本类别的完整说明均在Excel/CSV中。本文件先给最常用层级。\n\n## 已有解剖细类\n\n'
+ anatomy=c[c.namespace.eq('known_fine_anatomy')][['label','n_cells','biological_context_cn']]
+ doc+=table(anatomy,list(anatomy.columns))+'\n\n## 稳定8候选\n\n'
+ st=c[c.namespace.eq('stable8')][['label','n_cells','dominant_known_fine_type','dominant_side','type_side_purity','dominant_parent_coverage','correspondence_cn']]
+ doc+=table(st,list(st.columns))+'\n\n## 最细20单元：逐类对应与主要非KC伙伴\n\n'+table(fine,list(fine.columns))+'\n\n输入/输出top3为非KC伙伴计数，不是功能标签。全部4038条目还包括不同K/方法/种子、父类内细分、旧对照和残余；请用class_dictionary按namespace检索。所有类别均可从member_list_index找到CSV/TXT。\n'
+ (P/'类别说明与分析入口.md').write_text(doc)
+ notes=json.loads((P/'exports/slide_content.json').read_text());(P/'exports/逐页讲稿.md').write_text('# 吴老师组会：12页逐页讲稿\n\n'+'\n\n'.join(f"## {i+1:02} {r['title']}\n\n**结论：{r['summary']}**\n\n{r['notes']}" for i,r in enumerate(notes)))
+ # Browser: self-contained classifications with exact text root IDs and downloads.
+ records=[];roots=m.groupby('class_key').root_id.agg(list).to_dict()
+ for row in c.fillna('').to_dict('records'):row['roots']=roots[row['class_key']];records.append(row)
+ payload=json.dumps(records,ensure_ascii=False,separators=(',',':')).replace('</','<\\/')
+ html='''<!doctype html><meta charset="utf-8"><title>KC全量分类图谱</title><style>body{font:16px system-ui;color:#243746;margin:32px;max-width:1500px}h1{font-size:27px}input,select,button{padding:10px;font:inherit;margin:5px}table{border-collapse:collapse;width:100%;font-size:14px}td,th{border-bottom:1px solid #ddd;padding:8px;text-align:left}tr:hover{background:#eaf3f3;cursor:pointer}#detail{white-space:pre-wrap;background:#f2f6f6;padding:16px}textarea{width:95%;height:150px}small{color:#556b75}</style><h1>全部5,177 KC：分类、含义与逐类名单</h1><p>4038个方法版本条目；包含不同K、种子和残余，并非4038种生物类型。点击一类查看完整ID。离线运行。</p><select id="ns"></select><input id="q" placeholder="搜索标签、已知类型或说明"><span id="count"></span><div id="detail">选择下方任意类别。</div><table><thead><tr><th>namespace</th><th>类别</th><th>n</th><th>主导已知细类</th><th>侧别</th><th>纯度</th><th>覆盖</th><th>状态</th></tr></thead><tbody id="rows"></tbody></table><script>const data=PAYLOAD;const ns=document.querySelector('#ns'),q=document.querySelector('#q'),rows=document.querySelector('#rows'),detail=document.querySelector('#detail');const option=(s)=>{let o=document.createElement('option');o.textContent=s;ns.append(o)};option('finest20');[...new Set(data.map(x=>x.namespace))].filter(x=>x!='finest20').forEach(option);let selected;function render(){rows.replaceChildren();let subset=data.filter(r=>r.namespace==ns.value&&JSON.stringify(r).toLowerCase().includes(q.value.toLowerCase()));document.querySelector('#count').textContent=subset.length+'类';subset.forEach(r=>{let tr=document.createElement('tr');[r.namespace,r.label,r.n_cells,r.dominant_known_fine_type,r.dominant_side,(100*r.type_side_purity).toFixed(1)+'%',(100*r.dominant_parent_coverage).toFixed(1)+'%',r.evidence_status].forEach(v=>{let td=document.createElement('td');td.textContent=v;tr.append(td)});tr.onclick=()=>show(r);rows.append(tr)})}function show(r){selected=r;detail.replaceChildren();let p=document.createElement('p');p.textContent=r.class_key+' | n='+r.n_cells+'\n'+r.meaning_cn+'\n'+r.biological_context_cn+'；'+r.correspondence_cn+'\n主要非KC输入计数：'+r.in_top3_nonKC_partner_counts+'\n主要非KC输出计数：'+r.out_top3_nonKC_partner_counts+'\n'+r.recommended_analysis_cn;detail.append(p);let a=document.createElement('button');a.textContent='下载本类root_id.txt';a.onclick=()=>{let link=document.createElement('a');link.href=URL.createObjectURL(new Blob([r.roots.join('\\n')+'\\n'],{type:'text/plain;charset=utf-8'}));link.download='KC_'+r.label.replace(/[^a-zA-Z0-9_-]/g,'_')+'.txt';link.click();setTimeout(()=>URL.revokeObjectURL(link.href),1000)};detail.append(a);let area=document.createElement('textarea');area.value=r.roots.join('\\n');area.readOnly=true;detail.append(area)}ns.onchange=render;q.oninput=render;render();</script>'''.replace('PAYLOAD',payload)
+ # Convert intended JS newlines to escaped literals (the HTML source is not shell text).
+ html=html.replace("+'\n'+","+'\\n'+").replace("+'\n主要","+'\\n主要").replace("+'\n'+r.","+'\\n'+r.")
+ (P/'exports/KC_atlas_browser.html').write_text(html)
+ print('Collaborator README, biological dictionary, slide notes and offline browser generated.')
+if __name__=='__main__':main()

wu_20261109_full_atlas_briefing/code/download_data.py

@@ -0,0 +1,27 @@
+"""Download the complete raw KC subset and classification replay sources.
+Default: two complete current raw archives (~1.02 GB). Optionally materialize
+all previous publicly archived objects (~19.7 GB), pinned to their revisions.
+Read the provenance/license notes before reuse; no single blanket licence.
+"""
+from pathlib import Path
+import argparse,json,hashlib
+from huggingface_hub import hf_hub_download
+P=Path(__file__).resolve().parents[1];REPO='cyd0806/neuronmimic-data'
+def digest(path):
+ h=hashlib.sha256()
+ with open(path,'rb') as f:
+  for chunk in iter(lambda:f.read(8*1024*1024),b''):h.update(chunk)
+ return h.hexdigest()
+def main():
+ ap=argparse.ArgumentParser();ap.add_argument('--out',type=Path,default=Path('downloaded_data'));ap.add_argument('--all-historical',action='store_true');a=ap.parse_args()
+ receipt_path=P/'metadata/publication_receipt.json'
+ if not receipt_path.exists():receipt_path=Path(hf_hub_download(REPO,'wu_20261109_full_atlas_briefing/metadata/publication_receipt.json',repo_type='dataset'))
+ receipt=json.loads(receipt_path.read_text());rev=receipt['payload_commit']
+ for row in json.loads((P/'metadata/raw_archives.json').read_text()):
+  f=hf_hub_download(REPO,'wu_20261109_full_atlas_briefing/'+row['path'],repo_type='dataset',revision=rev,local_dir=a.out)
+  assert Path(f).stat().st_size==row['bytes'] and digest(f)==row['sha256'];print('Verified',row['path'])
+ if a.all_historical:
+  for row in json.loads((P/'data/previous_public_data_objects.json').read_text()):
+   f=hf_hub_download(REPO,row['path'],repo_type='dataset',revision=row['revision'],local_dir=a.out);assert Path(f).stat().st_size==row['size']
+  print('Previous immutable objects downloaded; HF validates transport blobs and LFS content.')
+if __name__=='__main__':main()

wu_20261109_full_atlas_briefing/code/final_checks.py

@@ -0,0 +1,22 @@
+from pathlib import Path
+import hashlib,json,zipfile,pandas as pd,openpyxl
+P=Path(__file__).resolve().parents[1]
+def main():
+ c=pd.read_parquet(P/'data/KC5177_master.parquet');m=pd.read_parquet(P/'data/all_class_memberships.parquet');d=pd.read_csv(P/'data/class_dictionary.csv');summary=json.loads((P/'metadata/atlas_summary.json').read_text())
+ assert len(c)==c.root_id.nunique()==5177
+ assert c.root_id.map(lambda s:isinstance(s,str) and s.isdecimal() and len(s)==18).all()
+ sim=c[c.simulation_gid.ne('')].copy();assert len(sim)==32 and sorted(sim.simulation_gid.astype(int))==list(range(32));assert sim.original61.eq('abp_R_K3_C0').all()
+ assert c.finest20.eq('UNASSIGNED').sum()==3052
+ assert d.class_key.nunique()==4038 and len(m)==230899
+ assert m.groupby('class_key').size().reindex(d.class_key).to_numpy().tolist()==d.n_cells.tolist()
+ raw=json.loads((P/'metadata/raw_archives.json').read_text());assert sum(x['bytes'] for x in raw)==1023635799
+ for name in ['atlas_validation','isolated_atlas_replay','briefing_validation','manuscript_validation','browser_logic_check']:
+  assert json.loads((P/'validation'/f'{name}.json').read_text())['status']=='PASS'
+ w=openpyxl.load_workbook(P/'exports/KC_all_classifications_5177.xlsx',read_only=True);rows=w['KC5177'].iter_rows();heads=[x.value for x in next(rows)];ix=heads.index('root_id');ids=[r[ix] for r in rows];assert all(x.data_type=='s' for x in ids);assert [x.value for x in ids]==c.root_id.tolist()
+ with zipfile.ZipFile(P/'exports/KC_all_classifications_handoff.zip') as z:
+  assert z.testzip() is None and sum(n.startswith('member_lists/') for n in z.namelist())==8076
+  assert z.read('data/class_dictionary.csv')==(P/'data/class_dictionary.csv').read_bytes()
+  assert z.read('KC_all_classifications_5177.xlsx')==(P/'exports/KC_all_classifications_5177.xlsx').read_bytes()
+ report=dict(status='PASS',all_5177_cells_retained=True,all_namespace_memberships_conserved=True,simulation_order_original32=True,all_8076_lists_packaged=True,Excel_ID_text_after_final_seal=True,raw_archives_bytes=sum(x['bytes'] for x in raw),classification_replay_byte_identical=True,article_and_slides_checked=True)
+ (P/'validation/final_checks.json').write_text(json.dumps(report,indent=2)+'\n');print(json.dumps(report))
+if __name__=='__main__':main()

wu_20261109_full_atlas_briefing/code/journal_layout.py

@@ -0,0 +1,38 @@
+"""Keep six main figures, ten Extended Data figures and full supplementary art."""
+import re
+
+EXTENDED=[1,2,3,4,5,13,15,16,22,23]
+SUPPLEMENTARY=[i for i in range(1,25) if i not in EXTENDED]
+
+def records():
+    rows=[dict(source=f'Figure{i}',name=f'Figure{i}',section='main',label=f'Figure {i}') for i in range(1,7)]
+    rows += [dict(source=f'ExtendedData{old}',name=f'ExtendedData{i}',section='extended',
+                  label=f'Extended Data Figure {i}') for i,old in enumerate(EXTENDED,1)]
+    rows += [dict(source=f'ExtendedData{old}',name=f'SupplementaryFigure{i}',section='supplementary',
+                  label=f'Supplementary Figure {i}') for i,old in enumerate(SUPPLEMENTARY,1)]
+    return rows
+
+def convert_citations(text):
+    mapping={old:('Extended Data',i) for i,old in enumerate(EXTENDED,1)}
+    mapping.update({old:('Supplementary',i) for i,old in enumerate(SUPPLEMENTARY,1)})
+    component=r'\d+(?:[a-z](?:[–-][a-z])?)?(?:\s*[–-]\s*\d+[a-z]?)?'
+    pattern=re.compile(r'Extended Data Figs?\.\s*'+component+r'(?:(?:,\s*| and )(?:'+component+r'|[a-z]))*')
+    def replace(match):
+        content=re.sub(r'^Extended Data Figs?\.\s*','',match.group())
+        # Expand numerical ranges before preserving panel-letter groups.
+        content=re.sub(r'(\d+)\s*[–-]\s*(\d+)',lambda m:', '.join(str(i) for i in range(int(m[1]),int(m[2])+1)),content)
+        parts=re.findall(r'(\d+)([a-z](?:[–-][a-z])?(?:,\s*[a-z])*)?',content)
+        groups=[]
+        for old,suffix in parts:
+            family,n=mapping[int(old)]
+            ref=str(n)+suffix.replace(' ','')
+            if groups and groups[-1][0]==family:
+                groups[-1][1].append(ref)
+            else:
+                groups.append((family,[ref]))
+        formatted=[]
+        for family,refs in groups:
+            values=', '.join(refs[:-1])+' and '+refs[-1] if len(refs)>1 else refs[0]
+            formatted.append(family+(' Figs. ' if len(refs)>1 else ' Fig. ')+values)
+        return ' and '.join(formatted)
+    return pattern.sub(replace,text)

wu_20261109_full_atlas_briefing/code/package_raw.py

@@ -0,0 +1,38 @@
+"""Archive all raw KC atlas objects and all classification replay sources.
+Large archives are published to HF, not committed to GitHub.
+"""
+from pathlib import Path
+import json,hashlib,zipfile,shutil
+from build_atlas import sha,P,R
+OUT=R/'deliveries/wu_20261109_full_atlas_raw'
+def main():
+ OUT.mkdir(exist_ok=True)
+ archive=R/'deliveries/kc_all5177_structural_atlas_20261030/KC_all5177_complete_structural_atlas_20261030.zip'
+ with zipfile.ZipFile(archive) as z:
+  assert z.testzip() is None
+  names=z.namelist();geom=[n for n in names if '/geometry/' in n and n.endswith('.json')];assert len(geom)==5177,len(geom)
+ raw=dict(local=str(archive.relative_to(R)),path='raw/'+archive.name,bytes=archive.stat().st_size,sha256=sha(archive),archive_members=len(names),geometry_objects=len(geom),scope='All 5177 KC geometry, incident synapses, endpoints, directed pairs, branch profiles, annotations, analysis, source and audits; FlyWire v783 subset.')
+ paths=[R/'artifacts/wu_20261026_multiscale_units',R/'artifacts/wu_20261030_fine_units_integrated_atlas',R/'projects/wu_20261102_followup_20261103']
+ paths += [p for p in sorted((R/'artifacts').glob('*')) if p.name in ['wu_20261026_density_refinement','wu_20261027_branch_contact_model','wu_20261027_local_tree_refinement','wu_20261028_open_tools','wu_20261028_parent_free_model']]
+ paths += [R/'artifacts/nature_kc_20261102/spatial_explanation_v1']
+ target=OUT/'KC_classification_full_sources.zip';manifest=[]
+ with zipfile.ZipFile(target,'w',zipfile.ZIP_DEFLATED) as z:
+  for base in paths:
+   for f in sorted(base.rglob('*')):
+    if f.is_file() and '__pycache__' not in f.parts:
+     z.write(f,str(f.relative_to(R)));manifest.append(dict(path=str(f.relative_to(R)),bytes=f.stat().st_size,sha256=sha(f)))
+  for f in [R/'simulation/scripts/build_fine_units_integrated_atlas.py',R/'20261102_吴老师组会纪要.pdf']:
+   z.write(f,str(f.relative_to(R)));manifest.append(dict(path=str(f.relative_to(R)),bytes=f.stat().st_size,sha256=sha(f)))
+  z.writestr('SOURCE_FILE_MANIFEST.json',json.dumps(manifest,ensure_ascii=False,indent=2))
+ with zipfile.ZipFile(target) as z:assert z.testzip() is None
+ second=dict(local=str(target.relative_to(R)),path='raw/'+target.name,bytes=target.stat().st_size,sha256=sha(target),source_files=len(manifest),scope='Complete multiscale discovery/confirmation labels and protocols, full joint features and assignments, all parent/focal refits, integrated finest atlas, historical method controls and source scripts.')
+ (P/'metadata/raw_archives.json').write_text(json.dumps([raw,second],ensure_ascii=False,indent=2)+'\n')
+ # Every existing remote object is indexed with an immutable revision for full historical replay.
+ inv=json.load(open('/tmp/wu_hf_inventory.json'));rows=[]
+ for f in inv['files']:
+  if f['path']=='.gitattributes':continue
+  if f['path'].split('/')[0].startswith(('wu_','structure_','flywire_','p1','p0','physics_','provenance')):
+   rows.append(dict(**f,revision=inv['revision'],url=f"https://huggingface.co/datasets/cyd0806/neuronmimic-data/resolve/{inv['revision']}/{f['path']}"))
+ (P/'data/previous_public_data_objects.json').write_text(json.dumps(rows,ensure_ascii=False,indent=2)+'\n')
+ print(json.dumps(dict(raw_archives=[raw,second],previous_objects_indexed=len(rows),previous_bytes=sum(x['size'] for x in rows)),ensure_ascii=False,indent=2))
+if __name__=='__main__':main()

wu_20261109_full_atlas_briefing/code/publish_release.py

@@ -0,0 +1,68 @@
+"""Additive publication of the reviewed atlas, article and 12-slide briefing.
+GitHub receives code and compact reviewable data. HF also receives both full
+raw archives. No old prefix is overwritten; each remote object is verified.
+"""
+from pathlib import Path
+import hashlib,json,re,subprocess
+import httpx
+from huggingface_hub import HfApi,CommitOperationAdd,set_client_factory
+P=Path(__file__).resolve().parents[1];R=P.parents[1];W=R/'deliveries/wu_structure_function_code_20261105'
+REPO='cyd0806/neuronmimic-data';PREFIX=P.name
+
+def files():
+ return [f for f in sorted(P.rglob('*')) if f.is_file() and '__pycache__' not in f.parts and f.name!='publication_receipt.json' and not f.is_relative_to(P/'exports/overleaf') and f.suffix not in ['.aux','.bbl','.blg','.log','.out']]
+def identity(obj):return obj.size,obj.blob_id,getattr(getattr(obj,'lfs',None),'sha256',None)
+def sha(p):
+ h=hashlib.sha256()
+ with p.open('rb') as f:
+  for b in iter(lambda:f.read(8*1024*1024),b''):h.update(b)
+ return h.hexdigest()
+def main():
+ for f in ['atlas_validation.json','isolated_atlas_replay.json','briefing_validation.json','manuscript_validation.json','final_checks.json']:
+  assert json.loads((P/'validation'/f).read_text())['status']=='PASS',f
+ assert json.loads((P/'validation/visual_review.json').read_text())['status']=='PASS'
+ head=subprocess.check_output(['git','rev-parse','HEAD'],cwd=W,text=True).strip()
+ remote=subprocess.check_output(['git','ls-remote','origin','refs/heads/research/wu-structure-function-20261104'],cwd=W,text=True).split()[0]
+ assert head==remote and not subprocess.check_output(['git','status','--porcelain','-z'],cwd=W)
+ payload=files();pat=re.compile(rb'(?:gh[pousr]_[A-Za-z0-9_]{20,}|github_pat_[A-Za-z0-9_]{20,}|hf_[A-Za-z0-9]{20,})')
+ tracked=set(subprocess.check_output(['git','ls-files','-z'],cwd=W,text=True).split('\0'))
+ for f in payload:
+  assert not pat.search(f.read_bytes()),f.name
+  path='projects/'+P.name+'/'+str(f.relative_to(P));assert path in tracked,path
+  assert (W/path).read_bytes()==f.read_bytes(),path
+ for line in (P/'SHA256SUMS').read_text().splitlines():
+  d,n=line.split('  ',1);assert sha(P/n)==d,n
+ proxy=subprocess.run(['git','config','--get','https.proxy'],capture_output=True,text=True).stdout.strip()
+ set_client_factory(lambda:httpx.Client(proxy=proxy or None,timeout=120,follow_redirects=True))
+ api=HfApi();assert api.whoami()['name']=='cyd0806'
+ before=api.repo_info(REPO,repo_type='dataset').sha
+ old={o.path:identity(o) for o in api.list_repo_tree(REPO,repo_type='dataset',revision=before,recursive=True) if hasattr(o,'size')}
+ assert not any(n.startswith(PREFIX+'/') for n in old)
+ items=[(str(f.relative_to(P)),f) for f in payload]
+ for row in json.loads((P/'metadata/raw_archives.json').read_text()):
+  f=R/row['local'];assert f.stat().st_size==row['bytes'] and sha(f)==row['sha256'];items.append((row['path'],f))
+ operations=[CommitOperationAdd(path_in_repo=PREFIX+'/'+name,path_or_fileobj=f) for name,f in items]
+ print('Publishing',len(items),'objects;',sum(f.stat().st_size for _,f in items),'bytes',flush=True)
+ commit=api.create_commit(REPO,repo_type='dataset',parent_commit=before,operations=operations,commit_message='Release all 5177 KC classifications, full raw archives, class dictionary, article and 12-slide Wu briefing',num_threads=4)
+ current={o.path:o for o in api.list_repo_tree(REPO,repo_type='dataset',revision=commit.oid,recursive=True) if hasattr(o,'size')}
+ checked=[]
+ for name,f in items:
+  obj=current[PREFIX+'/'+name];digest=sha(f);assert obj.size==f.stat().st_size,name
+  if identity(obj)[2]:assert identity(obj)[2]==digest,name
+  else:
+   b=f.read_bytes();assert obj.blob_id==hashlib.sha1(b'blob '+str(len(b)).encode()+b'\0'+b).hexdigest(),name
+  checked.append(dict(path=name,bytes=f.stat().st_size,sha256=digest))
+ for name,ident in old.items():
+  if name!='.gitattributes':assert identity(current[name])==ident,name
+ if '.gitattributes' in old and identity(current['.gitattributes'])!=old['.gitattributes']:
+  from huggingface_hub import hf_hub_download
+  lines=lambda rev:set(Path(hf_hub_download(REPO,'.gitattributes',repo_type='dataset',revision=rev)).read_text().splitlines())
+  assert lines(before)<=lines(commit.oid)
+ receipt=dict(status='PUBLISHED_AND_VERIFIED',github_scientific_commit=head,branch='research/wu-structure-function-20261104',dataset=REPO,prefix=PREFIX,payload_commit=commit.oid,previous_dataset_revision=before,historical_objects_preserved=len(old)-int('.gitattributes' in old),summary=json.loads((P/'metadata/atlas_summary.json').read_text()),verified_files=checked)
+ body=json.dumps(receipt,ensure_ascii=False,indent=2).encode()
+ rec=api.create_commit(REPO,repo_type='dataset',parent_commit=commit.oid,operations=[CommitOperationAdd(path_in_repo=PREFIX+'/metadata/publication_receipt.json',path_or_fileobj=body)],commit_message='Record full atlas release hashes and preserved historical objects')
+ obj=api.get_paths_info(REPO,[PREFIX+'/metadata/publication_receipt.json'],repo_type='dataset',revision=rec.oid)[0]
+ assert obj.size==len(body) and obj.blob_id==hashlib.sha1(b'blob '+str(len(body)).encode()+b'\0'+body).hexdigest()
+ receipt['receipt_commit']=rec.oid;(P/'metadata/publication_receipt.json').write_text(json.dumps(receipt,ensure_ascii=False,indent=2)+'\n')
+ print(json.dumps({k:v for k,v in receipt.items() if k not in ['verified_files','summary']},indent=2))
+if __name__=='__main__':main()

wu_20261109_full_atlas_briefing/code/query_class.py

@@ -0,0 +1,20 @@
+"""Exact root-ID lookup; no credential or network required.
+Example: python code/query_class.py --namespace finest20 --label SU035
+         python code/query_class.py --root-id 720575940606579465
+"""
+from pathlib import Path
+import argparse,pandas as pd
+P=Path(__file__).resolve().parents[1]
+def main():
+ p=argparse.ArgumentParser();p.add_argument('--namespace');p.add_argument('--label');p.add_argument('--root-id');p.add_argument('--out',type=Path);a=p.parse_args()
+ m=pd.read_parquet(P/'data/all_class_memberships.parquet');d=pd.read_csv(P/'data/class_dictionary.csv',dtype={'label':str})
+ if a.root_id:
+  result=m[m.root_id.eq(a.root_id)].merge(d,on=['class_key','namespace','label'],validate='many_to_one');print(result[['namespace','label','n_cells','evidence_status','dominant_known_fine_type','dominant_side']].to_string(index=False))
+ else:
+  assert a.namespace,'Choose --root-id OR --namespace [--label]'
+  q=m[m.namespace.eq(a.namespace)]
+  if a.label:q=q[q.label.eq(a.label)]
+  assert len(q),'No matching class. See data/namespace_dictionary.csv and data/class_dictionary.csv'
+  result=q.merge(pd.read_parquet(P/'data/KC5177_master.parquet'),on='root_id',validate='many_to_one');print(result[['root_id','label','source_hemibrain_type','official_hemisphere','soma_x_um','soma_y_um','soma_z_um']].to_string(index=False))
+ if a.out:result.to_csv(a.out,index=False)
+if __name__=='__main__':main()

wu_20261109_full_atlas_briefing/code/seal_handoff.py

@@ -0,0 +1,34 @@
+"""Refresh the portable handoff after final documents and source tables."""
+from pathlib import Path
+import zipfile,shutil,tempfile,json
+from openpyxl import load_workbook
+P=Path(__file__).resolve().parents[1]
+def main():
+ x=P/'exports/KC_all_classifications_5177.xlsx';w=load_workbook(x)
+ if 'README' in w:del w['README']
+ s=w.create_sheet('README',0)
+ for row in [
+ ['内容','全部5,177 KC；各分类视图并存，详见项目README与类别说明'],
+ ['分类条目','4,038为不同方法/K/种子/作用域条目，不是生物类型数'],
+ ['root_id','18位文本，不得先转浮点数'],
+ ['known_fine','source_hemibrain_type保留细KCab-s/m/c；official原字段另存'],
+ ['unassigned','finest20未覆盖仍保留；original61之外为范围外，不是没功能'],
+ ['嵌套','34确认单元允许嵌套，不能把成员数相加算唯一细胞'],
+ ['比例','输入/输出分别使用全部同方向端点，包括KC接点和未知伙伴'],
+ ['完整名单','zip内member_list_index.csv对应8,076份CSV/TXT'],
+ ['原始数据','同一HF前缀raw/含完整结构与完整分类重放包'],
+ ['仿真编号','simulation_gid仅原32组对应0至31，其他细胞留空'],
+ ['功能解释','解剖对应与算法稳定性不能直接当作新的生理功能类型']]:s.append(row)
+ s.column_dimensions['A'].width=20;s.column_dimensions['B'].width=115;w.save(x)
+ old=P/'exports/KC_all_classifications_handoff.zip';temp=old.with_suffix('.new.zip')
+ with zipfile.ZipFile(old) as src,zipfile.ZipFile(temp,'w',zipfile.ZIP_DEFLATED) as out:
+  for info in src.infolist():
+   if info.filename.startswith('member_lists/') or info.filename=='member_list_index.csv':out.writestr(info.filename,src.read(info))
+  for file in sorted((P/'data').glob('*')):out.write(file,'data/'+file.name)
+  for file in [P/'README.md',P/'类别说明与分析入口.md',x,P/'metadata/raw_archives.json',P/'metadata/source_manifest.json',P/'metadata/integration_protocol.json']:
+   out.write(file,str(file.relative_to(P)) if file!=x else file.name)
+  for file in ['query_class.py','download_data.py','unpack_raw.py']:out.write(P/'code'/file,'code/'+file)
+  out.write(P/'exports/KC_atlas_browser.html','KC_atlas_browser.html')
+ with zipfile.ZipFile(temp) as z:assert z.testzip() is None;count=sum(n.startswith('member_lists/') for n in z.namelist());assert count==8076
+ temp.replace(old);print('Sealed portable handoff:',old.stat().st_size,'bytes;',count,'per-class lists')
+if __name__=='__main__':main()

wu_20261109_full_atlas_briefing/code/unpack_raw.py

@@ -0,0 +1,19 @@
+"""Map both portable ZIP layouts to the source tree expected by build_atlas.py."""
+from pathlib import Path
+import argparse,zipfile
+P=Path(__file__).resolve().parents[1];R=P.parents[1]
+def main():
+ a=argparse.ArgumentParser();a.add_argument('atlas_zip',type=Path);a.add_argument('classification_zip',type=Path);a.add_argument('--root',type=Path,default=R);q=a.parse_args()
+ for f,atlas in [(q.atlas_zip,True),(q.classification_zip,False)]:
+  with zipfile.ZipFile(f) as z:
+   for name in z.namelist():
+    if name.endswith('/'):continue
+    relative=Path(*Path(name).parts[1:]) if atlas else Path(name)
+    if atlas:relative=Path('artifacts/wu_20261030_all5177_structural_atlas')/relative
+    dest=q.root/relative
+    assert dest.resolve().is_relative_to(q.root.resolve()),'Unsafe archive path'
+    dest.parent.mkdir(parents=True,exist_ok=True)
+    with z.open(name) as src,dest.open('wb') as out:
+     import shutil;shutil.copyfileobj(src,out)
+ print('Raw sources arranged at',q.root)
+if __name__=='__main__':main()

wu_20261109_full_atlas_briefing/code/update_article.py

@@ -0,0 +1,37 @@
+"""Create an atlas-integrated article successor without modifying frozen studies."""
+from pathlib import Path
+import json,shutil
+P=Path(__file__).resolve().parents[1];R=P.parents[1];A=R/'projects/wu_20261108_active_refinement'
+def main():
+ summary=json.loads((P/'metadata/atlas_summary.json').read_text())
+ s=(A/'manuscript/article.md').read_text()
+ marker='We evaluated whether joint partner identity–position features improve prediction'
+ paragraph="""A complete classification crosswalk places these objects in a common anatomical hierarchy (Supplementary Fig. 14). The 52 historical candidate bins cover all 5,177 cells; eight cross-scale stable candidates contain 965 cells. A separate fixed-scale confirmation protocol retained 34 nested units with a union of 2,505 cells. Its twenty finest non-overlapping units cover 2,125 cells, while the other 3,052 remain explicitly represented in the registry. Ten of these twenty units lie entirely within one known fine anatomical type and hemisphere, and two exactly recover the bilateral KCab-p type-by-side classes. The remaining units have mixed fine-type memberships, with dominant type-by-side purity ranging down to 67.3%. Thus, structural stability, anatomical correspondence and subdivision depth provide complementary descriptions of the same population. Known-class recovery supplies interpretable anchors, while partial and mixed units nominate specific within-type and boundary comparisons.
+
+"""
+ assert marker in s;s=s.replace(marker,paragraph+marker,1)
+ method="""### Complete classification crosswalk and directed fingerprints
+
+We integrated the immutable 5,177-cell structural registry with the 52 historical candidates, eight stable candidates, nine cross-method consensus views, all forty selected scale-conditional sets (34 confirmed and six unconfirmed), and the twenty finest non-overlapping confirmed sets. We retained every original membership rather than projecting nested units onto a forced partition. The multiscale discovery and confirmation used eight fixed scales (K = 3, 9, 20, 30, 40, 50, 75 and 100), each with twenty initialization runs and twenty 20%-directed-pair-dropout runs in each batch. Stability S and coverage both had to reach 0.80 in initialization and dropout separately in discovery and confirmation. Confirmation is a within-connectome algorithmic stability test. Unconfirmed sets remain in the distributed tables.
+
+The crosswalk additionally retains all 23 full-population representation fits across K = 20, 40, 80 and 160, initialization and grid sensitivities, three complete within-known-parent refinements, thirty 61-cell local refits, and historical method controls. Labels are qualified by method, scale, seed and parent context. In total, 72 namespaces comprise 4,038 class entries and 230,899 cell-to-class membership records. These include multiple representations and residual groups of the same cells; the entry count is not a count of biological types. Unassigned states are explicit in the cell registry and namespace coverage table.
+
+For anatomical correspondence, we used the archived fine source annotation and hemisphere jointly. Each class records the dominant type-by-side count, its fraction of class members (purity), and its fraction of all cells of that known type and side (coverage). The original coarse official field and all source strings are retained. A single raw official alias for an alpha-prime/beta-prime ap1 cell explains why fine-source correspondence includes one cell omitted by the older normalized coarse-field ap1 denominator; the historical entry table is preserved unchanged. Root IDs remain decimal strings throughout CSV, Parquet and text-safe Excel export. The original 32/15/14 cell memberships and the 32-cell simulation order are unchanged.
+
+We counted directed endpoints by class, direction and partner cell type. Fractions use all endpoints of the same direction, including KC contacts and unannotated partners. Input ALPN, direct visual projection and APL fractions use all input endpoints; MBON output fractions use all output endpoints. The three highest-count non-KC partner types are also listed, with complete partner counts retained. These full-direction denominators differ from external-only fingerprints in the earlier multiscale archive. Centroids use the original soma coordinates in micrometres. These are descriptive anatomical profiles; nested classes share observations and are not treated as independent samples.
+
+"""
+ s=s.replace('## Methods\n\n','## Methods\n\n'+method,1)
+ s=s.replace('Source data and vector artwork for every main figure are available in the `wu_20261108_active_refinement` release', 'The complete classification crosswalk, all class member lists, raw structural atlas, classification replay archive and atlas-integrated article are available in the `wu_20261109_full_atlas_briefing` release. Source data and vector artwork for every main figure are preserved in the `wu_20261108_active_refinement` release')
+ s=s.replace('The manuscript release includes the compact inputs required to reproduce the new main figures independently of the large simulation archives.', 'The atlas successor includes code to rebuild all class profiles and vector atlas panels from the released raw data, together with the frozen dynamical source release and an independently compiling Overleaf package.')
+ (P/'manuscript/article.md').write_text(s)
+ legends=json.loads((A/'manuscript/legends.json').read_text())
+ legends['ExtendedData24']='Complete KC classification crosswalk and anatomical fingerprints. a, Unique-cell coverage for known fine anatomical labels, the complete historical partition, cross-scale stable candidates, the union of confirmed nested units, and the finest non-overlapping display set. Counts describe overlapping views of the same 5,177 cells and cannot be added across rows. b, Stable candidate coverage of the dominant known fine type and side, with class purity. Fine-source annotation retains the raw official ap1 alias and therefore differs by one cell from the historical normalized-field entry table for CM020. c, All twenty finest units, their cell counts, dominant known fine type and side, and purity: ten are pure type-by-side subsets and two recover complete known KCab-p classes. d, ALPN, APL and direct visual projection input fractions and MBON output fraction for every finest unit. Denominators are all same-direction endpoint records including KC contacts and unknown partners. Class memberships, unconfirmed units and all remaining cells are retained. This figure summarizes anatomical correspondence and algorithmic stability, not new physiological type assignments.'
+ (P/'manuscript/legends.json').write_text(json.dumps(legends,indent=2)+'\n')
+ layout=(A/'code/journal_layout.py').read_text().replace('range(1,24)','range(1,25)')
+ (P/'code/journal_layout.py').write_text(layout)
+ build=(A/'code/build_paper.py').read_text().replace('8 November 2026','9 November 2026').replace('Supplementary Figures 1--13','Supplementary Figures 1--14').replace('supplementary_figures=13','supplementary_figures=14')
+ (P/'code/build_paper.py').write_text(build)
+ shutil.copy2(A/'metadata/mesh_source_attribution.json',P/'metadata/mesh_source_attribution.json')
+ print('Article adds full-population class correspondence, complete method and Supplementary Fig. 14.')
+if __name__=='__main__':main()

wu_20261109_full_atlas_briefing/data/KC5177_master.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d70331ca8e3cec431d379f96721a673ca706faca43c9d9bd605bb1a16570990e
+size 1257910

wu_20261109_full_atlas_briefing/data/active_pair_summary.csv

@@ -0,0 +1,13 @@
+seed,mask,stage,original_count,reassigned_count,count_change,absolute_cell_count_change,common_acquisition_count_change,cell_count_changes,original_responding,reassigned_responding,response_status_changes,paired_onsets,maxabs_first_peak_change_ms,max_soma_RMS_assignment_mV
+20261109,2,fine,160,157,-3,3,-3,1,32,32,0,32,10.349999999999994,12.713817545768887
+20261109,2,main,158,157,-1,1,-1,1,32,32,0,32,11.099999999999994,13.26302674749731
+20261109,4,fine,125,190,65,65,65,32,32,32,0,32,2.162499999999966,10.243729108556785
+20261109,4,main,188,158,-30,30,-30,30,32,32,0,32,3.2999999999999545,9.743736034957461
+20261110,2,fine,157,158,1,3,1,2,32,32,0,32,31.68750000000003,9.449405324214943
+20261110,2,main,127,158,31,31,31,31,32,32,0,32,32.19999999999999,11.070103838071004
+20261110,4,fine,93,186,93,93,93,32,31,32,1,31,13.225000000000023,13.152969904811838
+20261110,4,main,188,216,28,30,28,30,32,32,0,32,210.75,16.77204642440717
+20261111,2,fine,156,158,2,4,2,4,32,32,0,32,19.037500000000023,9.713898334087816
+20261111,2,main,187,128,-59,63,-59,31,32,32,0,32,18.525000000000034,14.1313484765833
+20261111,4,fine,157,187,30,32,30,31,32,32,0,32,51.80000000000001,13.736395419930219
+20261111,4,main,159,188,29,31,29,31,32,32,0,32,56.92499999999998,14.337999680069034

wu_20261109_full_atlas_briefing/data/all_class_memberships.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9cfff5bcc92db1f7f16237ef9f0046b708e2ee1f8b7d33459a48a804ae5be7b3
+size 25483184

wu_20261109_full_atlas_briefing/data/all_class_memberships.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90cc3fb3199ba7a1be8c88eddfd8076f55221889fb45d36f3b3aea2a23557ffc
+size 484505

wu_20261109_full_atlas_briefing/data/all_class_partner_fingerprints.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e74a62bddbf9575a2eadca54a167d878177ad4374f54065489340d6aef00cb51
+size 6169762

wu_20261109_full_atlas_briefing/data/async_run_summary.csv

@@ -0,0 +1,18 @@
+stage,condition,seed,mask,n_waveforms,responding_cells,mean_soma_mV,first_input_ms,input_event_count,active_contacts,distinct_contact_trains,n_internal_arrivals,n_MBON_arrivals
+main,synchronous_engine_gate_m1,0,1,559,32,-51.88924595109778,100.0,1008,56,1,33037,2542
+main,contact_async_s20261109_m1,20261109,1,252,32,-59.81205501744324,100.196166,1008,56,56,15134,1154
+main,contact_async_s20261109_m2,20261109,2,158,32,-62.99585493278222,100.196504,2376,132,132,9583,711
+main,contact_async_s20261109_m4,20261109,4,188,32,-61.96453539987135,100.152811,3870,215,215,11432,863
+main,contact_async_s20261109_m7,20261109,7,281,32,-58.440649686254275,100.152811,7254,403,403,17106,1302
+main,contact_async_s20261110_m1,20261110,1,255,32,-59.56677455823665,100.154308,1008,56,56,15486,1159
+main,contact_async_s20261110_m2,20261110,2,127,32,-64.05475986644191,100.126097,2376,132,132,7742,576
+main,contact_async_s20261110_m4,20261110,4,188,32,-61.913357357983145,100.016413,3870,215,215,11393,854
+main,contact_async_s20261110_m7,20261110,7,280,32,-58.541043761012176,100.016413,7254,403,403,16957,1293
+main,contact_async_s20261111_m1,20261111,1,312,32,-57.63450425541075,100.020358,1008,56,56,18792,1433
+main,contact_async_s20261111_m2,20261111,2,187,32,-61.73887494110224,100.289072,2376,132,132,11489,859
+main,contact_async_s20261111_m4,20261111,4,159,32,-62.76703864915291,100.00526,3870,215,215,9649,711
+main,contact_async_s20261111_m7,20261111,7,312,32,-57.43858119146046,100.00526,7254,403,403,18921,1444
+fine,contact_async_s20261109_m1,20261109,1,250,32,-59.68553615818085,100.196166,1008,56,56,15128,1149
+fine,contact_async_s20261109_m7,20261109,7,281,32,-58.441015425529145,100.152811,7254,403,403,17106,1302
+fine,contact_async_s20261109_m2,20261109,2,160,32,-63.02505910444568,100.196504,2376,132,132,9623,711
+fine,contact_async_s20261109_m4,20261109,4,125,32,-63.993774356301046,100.152811,3870,215,215,7683,583

wu_20261109_full_atlas_briefing/data/centroid_runs.csv

@@ -0,0 +1,41 @@
+run,condition,stage,origin,seed,mask,scale,kind,R_nS,total_waveforms,responding_cells,direct_responding,indirect_responding,mean_soma_mV,recurrent_integrated_h_uS_ms,recurrent_abs_q_pC,scheduled_recurrent_arrivals,nonzero_recurrent_arrivals,input_events,first_input_ms
+main/centroid_s20261109_a0p0_R1_m2,centroid_s20261109_a0p0_R1_m2,main,new,20261109,2,0.0,centroid,1.0,0,0,0,0,-69.46775950830654,0.0,0.0,0,0,2376,103.471484
+main/centroid_s20261109_a0p0_R1_m4,centroid_s20261109_a0p0_R1_m4,main,new,20261109,4,0.0,centroid,1.0,0,0,0,0,-69.41933002269903,0.0,0.0,0,0,3870,106.445324
+main/centroid_s20261109_a0p5_R1_m2,centroid_s20261109_a0p5_R1_m2,main,new,20261109,2,0.5,centroid,1.0,0,0,0,0,-69.0053566372032,0.0,0.0,0,0,2376,101.99853
+main/centroid_s20261109_a0p5_R1_m4,centroid_s20261109_a0p5_R1_m4,main,new,20261109,4,0.5,centroid,1.0,0,0,0,0,-68.95835091733251,0.0,0.0,0,0,3870,103.476246
+main/centroid_s20261110_a0p0_R1_m2,centroid_s20261110_a0p0_R1_m2,main,new,20261110,2,0.0,centroid,1.0,0,0,0,0,-69.4677246964911,0.0,0.0,0,0,2376,105.450973
+main/centroid_s20261110_a0p0_R1_m4,centroid_s20261110_a0p0_R1_m4,main,new,20261110,4,0.0,centroid,1.0,0,0,0,0,-69.41925871496703,0.0,0.0,0,0,3870,105.381414
+main/centroid_s20261110_a0p5_R1_m2,centroid_s20261110_a0p5_R1_m2,main,new,20261110,2,0.5,centroid,1.0,0,0,0,0,-68.99024341107106,0.0,0.0,0,0,2376,103.083375
+main/centroid_s20261110_a0p5_R1_m4,centroid_s20261110_a0p5_R1_m4,main,new,20261110,4,0.5,centroid,1.0,0,0,0,0,-68.96111695951028,0.0,0.0,0,0,3870,102.713777
+main/centroid_s20261111_a0p0_R1_m2,centroid_s20261111_a0p0_R1_m2,main,new,20261111,2,0.0,centroid,1.0,0,0,0,0,-69.4677814518399,0.0,0.0,0,0,2376,105.717859
+main/centroid_s20261111_a0p0_R1_m4,centroid_s20261111_a0p0_R1_m4,main,new,20261111,4,0.0,centroid,1.0,0,0,0,0,-69.4192434914242,0.0,0.0,0,0,3870,106.380484
+main/centroid_s20261111_a0p5_R1_m2,centroid_s20261111_a0p5_R1_m2,main,new,20261111,2,0.5,centroid,1.0,0,0,0,0,-69.00710257967737,0.0,0.0,0,0,2376,103.457137
+main/centroid_s20261111_a0p5_R1_m4,centroid_s20261111_a0p5_R1_m4,main,new,20261111,4,0.5,centroid,1.0,0,0,0,0,-68.97130474738924,0.0,0.0,0,0,3870,103.261692
+main/permuted_s20261109_a1p0_R1_m2,permuted_s20261109_a1p0_R1_m2,main,new,20261109,2,1.0,permuted,1.0,157,32,10,22,-62.76070694461403,19.272038415369824,379.66729205298327,9630,9630,2376,100.196504
+main/permuted_s20261109_a1p0_R1_m4,permuted_s20261109_a1p0_R1_m4,main,new,20261109,4,1.0,permuted,1.0,158,32,10,22,-62.91628224000634,19.292050916320292,367.6639163122038,9640,9640,3870,100.152811
+main/permuted_s20261110_a1p0_R1_m2,permuted_s20261110_a1p0_R1_m2,main,new,20261110,2,1.0,permuted,1.0,158,32,10,22,-62.88903200037667,19.097929657100032,373.527315579738,9543,9543,2376,100.126097
+main/permuted_s20261110_a1p0_R1_m4,permuted_s20261110_a1p0_R1_m4,main,new,20261110,4,1.0,permuted,1.0,216,32,10,22,-60.88136245155981,26.0902974892186,500.28363420116466,13037,13037,3870,100.016413
+main/permuted_s20261111_a1p0_R1_m2,permuted_s20261111_a1p0_R1_m2,main,new,20261111,2,1.0,permuted,1.0,128,32,10,22,-63.948759816530384,15.403621981630597,304.89331307612474,7697,7697,2376,100.289072
+main/permuted_s20261111_a1p0_R1_m4,permuted_s20261111_a1p0_R1_m4,main,new,20261111,4,1.0,permuted,1.0,188,32,10,22,-61.81069603875952,22.946333589888606,434.9742494766065,11466,11466,3870,100.00526
+main/centroid_s20261109_a0p0_R0_m2,centroid_s20261109_a0p0_R0_m2,main,new,20261109,2,0.0,centroid,0.0,0,0,0,0,-69.46775950830654,0.0,0.0,0,0,2376,103.471484
+main/centroid_s20261109_a0p0_R0_m4,centroid_s20261109_a0p0_R0_m4,main,new,20261109,4,0.0,centroid,0.0,0,0,0,0,-69.41933002269903,0.0,0.0,0,0,3870,106.445324
+fine/centroid_s20261109_a0p0_R1_m2,centroid_s20261109_a0p0_R1_m2,fine,new,20261109,2,0.0,centroid,1.0,0,0,0,0,-69.46828989454828,0.0,0.0,0,0,2376,103.471484
+fine/centroid_s20261109_a0p0_R1_m4,centroid_s20261109_a0p0_R1_m4,fine,new,20261109,4,0.0,centroid,1.0,0,0,0,0,-69.41966565295837,0.0,0.0,0,0,3870,106.445324
+main/reference_R1_s20261109_m1,reference_R1_s20261109_m1,main,legacy,20261109,1,1.0,original,1.0,252,32,10,22,-59.81205501744324,30.28691893854595,571.1071339697731,15134,15134,1008,100.196166
+main/reference_R1_s20261109_m2,reference_R1_s20261109_m2,main,legacy,20261109,2,1.0,original,1.0,158,32,10,22,-62.99585493278222,19.17797966090221,364.8634726728625,9583,9583,2376,100.196504
+main/reference_R1_s20261109_m4,reference_R1_s20261109_m4,main,legacy,20261109,4,1.0,original,1.0,188,32,10,22,-61.96453539987135,22.878291086656954,430.71969300252994,11432,11432,3870,100.152811
+main/reference_R1_s20261109_m7,reference_R1_s20261109_m7,main,legacy,20261109,7,1.0,original,1.0,281,32,10,22,-58.440649686254275,34.233384125990554,645.8153944429656,17106,17106,7254,100.152811
+main/reference_R1_s20261110_m1,reference_R1_s20261110_m1,main,legacy,20261110,1,1.0,original,1.0,255,32,10,22,-59.56677455823665,30.99135897200412,579.7424631099063,15486,15486,1008,100.154308
+main/reference_R1_s20261110_m2,reference_R1_s20261110_m2,main,legacy,20261110,2,1.0,original,1.0,127,32,10,22,-64.05475986644191,15.493678235908156,297.8427421560385,7742,7742,2376,100.126097
+main/reference_R1_s20261110_m4,reference_R1_s20261110_m4,main,legacy,20261110,4,1.0,original,1.0,188,32,10,22,-61.913357357983145,22.8002423329495,432.32873738538046,11393,11393,3870,100.016413
+main/reference_R1_s20261110_m7,reference_R1_s20261110_m7,main,legacy,20261110,7,1.0,original,1.0,280,32,10,22,-58.541043761012176,33.93519786182751,642.676665607192,16957,16957,7254,100.016413
+main/reference_R1_s20261111_m1,reference_R1_s20261111_m1,main,legacy,20261111,1,1.0,original,1.0,312,32,10,22,-57.63450425541075,37.60749178625076,707.0273723256628,18792,18792,1008,100.020358
+main/reference_R1_s20261111_m2,reference_R1_s20261111_m2,main,legacy,20261111,2,1.0,original,1.0,187,32,10,22,-61.73887494110224,22.992362342074973,439.80364371725517,11489,11489,2376,100.289072
+main/reference_R1_s20261111_m4,reference_R1_s20261111_m4,main,legacy,20261111,4,1.0,original,1.0,159,32,10,22,-62.76703864915291,19.31006216717585,371.2156105345787,9649,9649,3870,100.00526
+main/reference_R1_s20261111_m7,reference_R1_s20261111_m7,main,legacy,20261111,7,1.0,original,1.0,312,32,10,22,-57.43858119146046,37.8656530485124,709.1772427121707,18921,18921,7254,100.00526
+main/reference_R0_s20261109_m1,reference_R0_s20261109_m1,main,legacy,20261109,1,1.0,original,0.0,5,1,1,0,-68.74613950998533,0.0,0.0,733,0,1008,100.196166
+main/reference_R0_s20261109_m2,reference_R0_s20261109_m2,main,legacy,20261109,2,1.0,original,0.0,2,1,1,0,-68.6680170847748,0.0,0.0,371,0,2376,100.196504
+main/reference_R0_s20261109_m4,reference_R0_s20261109_m4,main,legacy,20261109,4,1.0,original,0.0,2,1,1,0,-68.62984741484924,0.0,0.0,405,0,3870,100.152811
+main/reference_R0_s20261109_m7,reference_R0_s20261109_m7,main,legacy,20261109,7,1.0,original,0.0,1,1,1,0,-68.57195762146469,0.0,0.0,672,0,7254,100.152811
+main/reference_sync_m2,reference_sync_m2,main,legacy,0,2,0.0,original,1.0,0,0,0,0,-69.46664239145063,0.0,0.0,0,0,2376,100.0
+main/reference_sync_m4,reference_sync_m4,main,legacy,0,4,0.0,original,1.0,0,0,0,0,-69.41814503150125,0.0,0.0,0,0,3870,100.0

wu_20261109_full_atlas_briefing/data/class_dictionary.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80e9ff9fb4834f160b664decd937d02de9e1b21ea778cbbb8a1ebaed76f4896b
+size 512628

wu_20261109_full_atlas_briefing/data/dispersion_runs.csv

@@ -0,0 +1,41 @@
+run,condition,stage,origin,seed,mask,scale,R_nS,total_waveforms,responding_cells,direct_responding,indirect_responding,mean_soma_mV,recurrent_integrated_h_uS_ms,recurrent_abs_q_pC,scheduled_recurrent_arrivals,nonzero_recurrent_arrivals,input_events,first_input_ms
+main/async_R0_s20261110_m1,async_R0_s20261110_m1,main,new,20261110,1,1.0,0.0,2,1,1,0,-68.71842289502338,0.0,0.0,672,0,1008,100.154308
+main/async_R0_s20261110_m2,async_R0_s20261110_m2,main,new,20261110,2,1.0,0.0,2,1,1,0,-68.64850948637178,0.0,0.0,392,0,2376,100.126097
+main/async_R0_s20261110_m4,async_R0_s20261110_m4,main,new,20261110,4,1.0,0.0,1,1,1,0,-68.63136950784599,0.0,0.0,366,0,3870,100.016413
+main/async_R0_s20261110_m7,async_R0_s20261110_m7,main,new,20261110,7,1.0,0.0,1,1,1,0,-68.55955848172229,0.0,0.0,675,0,7254,100.016413
+main/async_R0_s20261111_m1,async_R0_s20261111_m1,main,new,20261111,1,1.0,0.0,3,1,1,0,-68.74401992645882,0.0,0.0,697,0,1008,100.020358
+main/async_R0_s20261111_m2,async_R0_s20261111_m2,main,new,20261111,2,1.0,0.0,1,1,1,0,-68.66362090038535,0.0,0.0,385,0,2376,100.289072
+main/async_R0_s20261111_m4,async_R0_s20261111_m4,main,new,20261111,4,1.0,0.0,1,1,1,0,-68.64031562077321,0.0,0.0,354,0,3870,100.00526
+main/async_R0_s20261111_m7,async_R0_s20261111_m7,main,new,20261111,7,1.0,0.0,1,1,1,0,-68.57186512436235,0.0,0.0,689,0,7254,100.00526
+main/dispersion_s20261109_a0p125_m2,dispersion_s20261109_a0p125_m2,main,new,20261109,2,0.125,1.0,0,0,0,0,-69.35601440557119,0.0,0.0,0,0,2376,100.024563
+main/dispersion_s20261109_a0p125_m4,dispersion_s20261109_a0p125_m4,main,new,20261109,4,0.125,1.0,0,0,0,0,-69.31109662242719,0.0,0.0,0,0,3870,100.019101
+main/dispersion_s20261109_a0p5_m2,dispersion_s20261109_a0p5_m2,main,new,20261109,2,0.5,1.0,0,0,0,0,-69.0056313745576,0.0,0.0,0,0,2376,100.098252
+main/dispersion_s20261109_a0p5_m4,dispersion_s20261109_a0p5_m4,main,new,20261109,4,0.5,1.0,0,0,0,0,-68.95929731387103,0.0,0.0,0,0,3870,100.076406
+main/dispersion_s20261110_a0p125_m2,dispersion_s20261110_a0p125_m2,main,new,20261110,2,0.125,1.0,0,0,0,0,-69.35186242608378,0.0,0.0,0,0,2376,100.015762
+main/dispersion_s20261110_a0p125_m4,dispersion_s20261110_a0p125_m4,main,new,20261110,4,0.125,1.0,0,0,0,0,-69.31300320533097,0.0,0.0,0,0,3870,100.002052
+main/dispersion_s20261110_a0p5_m2,dispersion_s20261110_a0p5_m2,main,new,20261110,2,0.5,1.0,0,0,0,0,-68.99018046640643,0.0,0.0,0,0,2376,100.063049
+main/dispersion_s20261110_a0p5_m4,dispersion_s20261110_a0p5_m4,main,new,20261110,4,0.5,1.0,0,0,0,0,-68.96172842725778,0.0,0.0,0,0,3870,100.008206
+main/dispersion_s20261111_a0p125_m2,dispersion_s20261111_a0p125_m2,main,new,20261111,2,0.125,1.0,0,0,0,0,-69.35563691326651,0.0,0.0,0,0,2376,100.036134
+main/dispersion_s20261111_a0p125_m4,dispersion_s20261111_a0p125_m4,main,new,20261111,4,0.125,1.0,0,0,0,0,-69.3168279174655,0.0,0.0,0,0,3870,100.000658
+main/dispersion_s20261111_a0p5_m2,dispersion_s20261111_a0p5_m2,main,new,20261111,2,0.5,1.0,0,0,0,0,-69.0070284437549,0.0,0.0,0,0,2376,100.144536
+main/dispersion_s20261111_a0p5_m4,dispersion_s20261111_a0p5_m4,main,new,20261111,4,0.5,1.0,0,0,0,0,-68.97217232284106,0.0,0.0,0,0,3870,100.00263
+fine/dispersion_s20261109_a0p5_m2,dispersion_s20261109_a0p5_m2,fine,new,20261109,2,0.5,1.0,0,0,0,0,-69.00643304693475,0.0,0.0,0,0,2376,100.098252
+fine/dispersion_s20261109_a0p5_m4,dispersion_s20261109_a0p5_m4,fine,new,20261109,4,0.5,1.0,0,0,0,0,-68.95997593775502,0.0,0.0,0,0,3870,100.076406
+main/reference_R1_s20261109_m1,reference_R1_s20261109_m1,main,legacy,20261109,1,1.0,1.0,252,32,10,22,-59.81205501744324,30.28691893854595,571.1071339697731,15134,15134,1008,100.196166
+main/reference_R1_s20261109_m2,reference_R1_s20261109_m2,main,legacy,20261109,2,1.0,1.0,158,32,10,22,-62.99585493278222,19.17797966090221,364.8634726728625,9583,9583,2376,100.196504
+main/reference_R1_s20261109_m4,reference_R1_s20261109_m4,main,legacy,20261109,4,1.0,1.0,188,32,10,22,-61.96453539987135,22.878291086656954,430.71969300252994,11432,11432,3870,100.152811
+main/reference_R1_s20261109_m7,reference_R1_s20261109_m7,main,legacy,20261109,7,1.0,1.0,281,32,10,22,-58.440649686254275,34.233384125990554,645.8153944429656,17106,17106,7254,100.152811
+main/reference_R1_s20261110_m1,reference_R1_s20261110_m1,main,legacy,20261110,1,1.0,1.0,255,32,10,22,-59.56677455823665,30.99135897200412,579.7424631099063,15486,15486,1008,100.154308
+main/reference_R1_s20261110_m2,reference_R1_s20261110_m2,main,legacy,20261110,2,1.0,1.0,127,32,10,22,-64.05475986644191,15.493678235908156,297.8427421560385,7742,7742,2376,100.126097
+main/reference_R1_s20261110_m4,reference_R1_s20261110_m4,main,legacy,20261110,4,1.0,1.0,188,32,10,22,-61.913357357983145,22.8002423329495,432.32873738538046,11393,11393,3870,100.016413
+main/reference_R1_s20261110_m7,reference_R1_s20261110_m7,main,legacy,20261110,7,1.0,1.0,280,32,10,22,-58.541043761012176,33.93519786182751,642.676665607192,16957,16957,7254,100.016413
+main/reference_R1_s20261111_m1,reference_R1_s20261111_m1,main,legacy,20261111,1,1.0,1.0,312,32,10,22,-57.63450425541075,37.60749178625076,707.0273723256628,18792,18792,1008,100.020358
+main/reference_R1_s20261111_m2,reference_R1_s20261111_m2,main,legacy,20261111,2,1.0,1.0,187,32,10,22,-61.73887494110224,22.992362342074973,439.80364371725517,11489,11489,2376,100.289072
+main/reference_R1_s20261111_m4,reference_R1_s20261111_m4,main,legacy,20261111,4,1.0,1.0,159,32,10,22,-62.76703864915291,19.31006216717585,371.2156105345787,9649,9649,3870,100.00526
+main/reference_R1_s20261111_m7,reference_R1_s20261111_m7,main,legacy,20261111,7,1.0,1.0,312,32,10,22,-57.43858119146046,37.8656530485124,709.1772427121707,18921,18921,7254,100.00526
+main/reference_R0_s20261109_m1,reference_R0_s20261109_m1,main,legacy,20261109,1,1.0,0.0,5,1,1,0,-68.74613950998533,0.0,0.0,733,0,1008,100.196166
+main/reference_R0_s20261109_m2,reference_R0_s20261109_m2,main,legacy,20261109,2,1.0,0.0,2,1,1,0,-68.6680170847748,0.0,0.0,371,0,2376,100.196504
+main/reference_R0_s20261109_m4,reference_R0_s20261109_m4,main,legacy,20261109,4,1.0,0.0,2,1,1,0,-68.62984741484924,0.0,0.0,405,0,3870,100.152811
+main/reference_R0_s20261109_m7,reference_R0_s20261109_m7,main,legacy,20261109,7,1.0,0.0,1,1,1,0,-68.57195762146469,0.0,0.0,672,0,7254,100.152811
+main/reference_sync_m2,reference_sync_m2,main,legacy,0,2,0.0,1.0,0,0,0,0,-69.46664239145063,0.0,0.0,0,0,2376,100.0
+main/reference_sync_m4,reference_sync_m4,main,legacy,0,4,0.0,1.0,0,0,0,0,-69.41814503150125,0.0,0.0,0,0,3870,100.0

wu_20261109_full_atlas_briefing/data/entry_roles_including_residuals.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c4349355151317951be2a15b89624d0cf34187ee7715be129761be700e69c34
+size 41327

wu_20261109_full_atlas_briefing/data/finest20_source_summary.csv

@@ -0,0 +1,21 @@
+unit_id,n_cells,confirmed_scales,evidence_status,evidence_tier,protocol_version,dominant_official_type,dominant_side,official_type_purity,original_stable8_overlap,legacy61_pass,original_cm_lineage,mapped_official_type,soma_centroid_x_um,soma_centroid_y_um,soma_centroid_z_um,soma_span_x_um,soma_span_y_um,soma_span_z_um,soma_pair_distance_q25_um,soma_pair_distance_median_um,soma_pair_distance_q75_um,n_input_endpoints,n_output_endpoints,n_unique_input_partners,n_unique_output_partners,n_internal_directed_pairs,internal_synapse_pairs,n_incoming_directed_pairs,n_outgoing_directed_pairs,input_alpn_fraction,input_visual_projection_fraction,input_apl_fraction,output_mbon_fraction,output_dan_fraction,output_mbin_fraction,input_mapping_error_median_um,output_mapping_error_median_um
+SU002,206,30,CONFIRMED_SCALE_CONDITIONAL,COMPUTED_MODEL_RESULT,wu_20261026_scale_conditional_v1,KCab,right,1.0,0,False,"{""CM007"": 206}",KCab,627.731708737864,132.6768932038835,184.76737864077677,113.12,77.888,57.0,25.797838667182532,51.52132172217633,68.67372271834392,43627,58154,1798,8473,10003,13809,21093,33185,0.3074930662204598,0.0002292158525683,0.1016113874435555,0.2220999415345462,0.1038449633731127,0.1165354059909894,0.1760089200348208,0.2204082441010488
+SU003,185,30,CONFIRMED_SCALE_CONDITIONAL,COMPUTED_MODEL_RESULT,wu_20261026_scale_conditional_v1,KCab,right,1.0,0,False,"{""CM008"": 180, ""CM041"": 5}",KCab,628.9972324324325,133.7596108108108,186.6722162162162,95.08799999999997,68.928,45.84,22.729637385579775,45.14944038637819,59.03566094381089,41555,50554,1698,7412,8291,11703,18192,28471,0.3537239802671159,0.0,0.0887979785826013,0.2102899869446532,0.1089330221149661,0.1296633303002729,0.1649411547827818,0.2198728789627961
+SU004,171,30,CONFIRMED_SCALE_CONDITIONAL,COMPUTED_MODEL_RESULT,wu_20261026_scale_conditional_v1,KCab,left,1.0,0,False,"{""CM009"": 171}",KCab,440.0809356725146,129.48327485380116,201.9031578947368,94.464,85.408,62.52000000000001,28.036119259202778,46.89825241946655,66.12789692243302,37437,49121,1780,7161,7530,11063,17728,26988,0.3092128108555707,0.0,0.1031065523412666,0.2278048085340282,0.0952138596526943,0.1318784226705482,0.175034648085797,0.2208103972183186
+SU005,168,30,CONFIRMED_SCALE_CONDITIONAL,COMPUTED_MODEL_RESULT,wu_20261026_scale_conditional_v1,KCab,right,1.0,0,False,"{""CM010"": 167, ""CM052"": 1}",KCab,631.5749047619048,134.38795238095238,186.9495238095238,93.34399999999994,70.112,43.360000000000014,23.42844303389553,51.10231415320549,66.01634268285298,35181,40659,1492,6694,6448,8855,14618,23426,0.400386572297547,0.0,0.0822318865296609,0.1832804545119161,0.1087582085147199,0.1583167318428884,0.1635115561525475,0.218377571133901
+SU011,144,30,CONFIRMED_SCALE_CONDITIONAL,COMPUTED_MODEL_RESULT,wu_20261026_scale_conditional_v1,KCab,left,1.0,0,False,"{""CM017"": 144}",KCab,439.1731944444444,135.1998888888889,203.61805555555557,72.89600000000002,74.368,54.16,27.549559860886816,54.32469785381528,64.37930947074997,29683,32635,1511,5270,5199,7196,12428,19172,0.4149850082538827,0.0,0.0800795067883974,0.1724222460548491,0.1055308717634441,0.1570706296920484,0.1634161785452942,0.2200937852633102
+SU013,132,40,CONFIRMED_SCALE_CONDITIONAL,COMPUTED_MODEL_RESULT,wu_20261026_scale_conditional_v1,KCg-d,left,0.9621212121212122,0,True,"{""CM018"": 132}",KCg-d,428.9747575757576,138.19254545454544,203.46303030303037,87.68,91.584,59.91999999999999,24.987864413779683,43.872671390989126,58.40057972464881,21756,32309,2293,8990,4123,5725,10077,20610,0.0130538701967273,0.2704541275969847,0.0956058098915241,0.1766071373301556,0.0675663127921012,0.1300875916927172,0.1686578254733719,0.2603061424596812
+SU017,116,50;75,CONFIRMED_SCALE_CONDITIONAL,COMPUTED_MODEL_RESULT,wu_20261026_scale_conditional_v1,KCapbp-m,right,1.0,116,False,"{""CM012"": 116}",KCapbp-m,624.8166896551725,123.76362068965516,181.26999999999995,126.476,77.44,65.16,24.876558258690302,61.28502919326016,77.91072899961875,18481,35853,1334,7398,4237,5892,10422,19592,0.1504247605649045,0.0,0.1406309182403549,0.1815189802805902,0.0919588318969124,0.2121161409087105,0.2290638832007591,0.3280866873689001
+SU018,108,100,CONFIRMED_SCALE_CONDITIONAL,COMPUTED_MODEL_RESULT,wu_20261026_scale_conditional_v1,KCapbp-ap1,right,0.9722222222222222,108,False,"{""CM020"": 108}",KCapbp-ap1,632.839888888889,131.18937037037037,176.61666666666665,115.84800000000008,87.16799999999999,68.6,27.147550724788605,68.78177562402624,85.93433733616743,13823,35954,1021,6912,3328,4295,7258,18170,0.2211531505461911,0.0007957751573464,0.1515589958764378,0.261584246537242,0.1513322578850753,0.1720531790621349,0.2448271425006528,0.3507579622422201
+SU019,104,75;100,CONFIRMED_SCALE_CONDITIONAL,COMPUTED_MODEL_RESULT,wu_20261026_scale_conditional_v1,KCapbp-ap1,left,1.0,104,False,"{""CM019"": 104}",KCapbp-ap1,434.5264230769232,128.1160384615385,200.40538461538455,108.89600000000002,86.304,70.44,36.64274855321366,60.28373420451578,79.3552702480155,15220,34900,1179,7137,3514,4645,7840,18340,0.2170170827858081,0.0,0.1526281208935611,0.2083667621776504,0.1432664756446991,0.1946704871060172,0.2326229291813807,0.333548548318031
+SU021,98,50;75,CONFIRMED_SCALE_CONDITIONAL,COMPUTED_MODEL_RESULT,wu_20261026_scale_conditional_v1,KCab,right,1.0,98,False,"{""CM016"": 98}",KCab,629.2470612244898,127.80208163265308,183.4942857142857,119.26800000000004,81.24799999999999,61.400000000000006,27.44296193926597,64.05335026366693,83.95816443920152,19782,29776,1141,5005,3862,6359,9518,16353,0.2645334142149428,0.0,0.0980689515721362,0.1937466415905427,0.1128425577646426,0.1190556152606125,0.1856510126369729,0.2251890837409445
+SU022,97,75;100,CONFIRMED_SCALE_CONDITIONAL,COMPUTED_MODEL_RESULT,wu_20261026_scale_conditional_v1,KCapbp-m,left,1.0,97,False,"{""CM014"": 97}",KCapbp-m,434.1399175257731,127.5481237113402,201.55628865979384,109.15200000000004,82.43199999999999,71.44,35.41896710578855,60.32466018035206,81.89516535417988,18588,33433,1374,6031,3755,5700,9571,16844,0.1520335700451904,0.0,0.1573057886808694,0.1877785421589447,0.1078275954894864,0.2138306463673616,0.2253166789504399,0.3229276102224262
+SU024,93,30,CONFIRMED_SCALE_CONDITIONAL,COMPUTED_MODEL_RESULT,wu_20261026_scale_conditional_v1,KCab,right,1.0,0,False,"{""CM026"": 88, ""CM043"": 5}",KCab,628.9441720430107,130.93475268817204,184.16903225806453,92.44799999999998,68.608,38.67999999999998,20.02208048994661,55.0922393617451,71.64063700429203,15865,19037,1068,3767,3108,4701,7014,11638,0.3743460447526001,0.0,0.0875512133627481,0.1323212691075274,0.1029048694647265,0.1530703367127173,0.1656950778625365,0.2147828542546746
+SU027,84,75,CONFIRMED_SCALE_CONDITIONAL,COMPUTED_MODEL_RESULT,wu_20261026_scale_conditional_v1,KCab,left,1.0,84,False,"{""CM022"": 84}",KCab,434.3397142857143,124.70280952380952,202.1357142857143,103.10399999999998,79.16399999999999,64.4,30.426956145624192,48.61856106881199,67.42597022801182,18068,25499,1156,4075,2865,5029,8141,13476,0.2885764888200133,0.0,0.1016714633606375,0.2143613475038236,0.0963959370955723,0.1405545315502568,0.1819939130461405,0.2256672190545586
+SU030,77,30,CONFIRMED_SCALE_CONDITIONAL,COMPUTED_MODEL_RESULT,wu_20261026_scale_conditional_v1,KCapbp-ap2,left,0.7922077922077922,0,False,"{""CM030"": 70, ""CM039"": 7}",KCapbp-ap2,429.6265454545455,127.3206753246753,205.00883116883116,92.88,71.52,68.4,29.30983010602548,47.10661427119706,68.24198467218798,11708,23561,1114,4837,2280,3434,6197,12375,0.148274683976768,0.0001708233686368,0.162282200204988,0.2324179788633759,0.1046220449047154,0.1897627435168286,0.2330980703304393,0.3376300603701052
+SU033,68,50,CONFIRMED_SCALE_CONDITIONAL,COMPUTED_MODEL_RESULT,wu_20261026_scale_conditional_v1,KCapbp-ap2,right,0.8529411764705882,0,False,"{""CM027"": 68}",KCapbp-ap2,632.0635882352941,127.54441176470588,179.29176470588234,109.25599999999996,82.656,63.56,30.637710799205887,62.65689546004897,77.49771896758887,8695,19369,897,4440,1527,2096,4840,10509,0.1774583093732029,0.0,0.150316273720529,0.2466828437193453,0.1051164231503949,0.1834890804894419,0.2323158082354929,0.34523561993675
+SU034,67,20;30;40;50;75;100,CONFIRMED_SCALE_CONDITIONAL,COMPUTED_MODEL_RESULT,wu_20261026_scale_conditional_v1,KCab-p,left,1.0,67,True,"{""CM031"": 67}",KCab-p,429.6287164179104,127.51420895522388,204.54149253731345,106.49599999999998,72.448,59.120000000000005,32.83829333982888,53.078926289065,69.12647420459386,17077,24933,948,3918,3455,7822,7161,11474,0.0055044797095508,0.0346079522164314,0.1056977220823329,0.1381301889062688,0.1488388882204307,0.1592668351181165,0.1888447248492856,0.2428397742039307
+SU035,61,20;30;40;50;75;100,CONFIRMED_SCALE_CONDITIONAL,COMPUTED_MODEL_RESULT,wu_20261026_scale_conditional_v1,KCab-p,right,1.0,61,True,"{""CM033"": 61}",KCab-p,629.0464918032789,126.8895081967213,182.3501639344262,121.15200000000004,82.016,56.88,22.044165172312223,65.29057072115248,81.7621190028158,15126,21556,1004,4190,2813,6476,6359,10816,0.0128255983075499,0.027568425228084,0.1053153510511701,0.106652440155873,0.1328168491371311,0.1537854889589905,0.1797952160907487,0.2391281172497233
+SU036,55,75,CONFIRMED_SCALE_CONDITIONAL,COMPUTED_MODEL_RESULT,wu_20261026_scale_conditional_v1,KCapbp-ap2,right,0.9090909090909092,0,False,"{""CM024"": 55}",KCapbp-ap2,628.9979636363636,124.89309090909089,179.75999999999996,121.70000000000005,72.51199999999999,58.56,33.17058576510219,62.034344036186944,81.03857846729545,7809,17654,916,4168,1066,1475,4634,9599,0.1488026635932897,0.0,0.1362530413625304,0.2240851931573581,0.1025263396397417,0.186473320493939,0.242539117438537,0.3406691243877582
+SU037,50,75,CONFIRMED_SCALE_CONDITIONAL,COMPUTED_MODEL_RESULT,wu_20261026_scale_conditional_v1,KCapbp-ap2,left,1.0,0,False,"{""CM025"": 50}",KCapbp-ap2,439.02344,127.8636,199.5288,100.48000000000002,76.06400000000001,72.0,31.345596692358548,59.25616808400625,80.1395778376702,8307,17131,904,3670,1067,1711,4417,8851,0.1593836523413988,0.0,0.1579390875165523,0.2177339326367404,0.1165722958379545,0.1813087385441597,0.2246458351443157,0.3402830272367351
+SU039,41,75,CONFIRMED_SCALE_CONDITIONAL,COMPUTED_MODEL_RESULT,wu_20261026_scale_conditional_v1,KCab,left,1.0,0,False,"{""CM029"": 41}",KCab,436.03687804878047,138.76341463414633,206.35121951219512,68.096,71.29599999999999,47.47999999999999,14.886275137617702,58.39563516480277,64.56445571829296,6583,7378,724,1747,864,1466,2908,4573,0.3624487315813459,0.0,0.0935743581953516,0.1204933586337761,0.1057197072377338,0.1416373000813228,0.1641497336189276,0.2144283664929621

wu_20261109_full_atlas_briefing/data/namespace_dictionary.csv

@@ -0,0 +1,73 @@
+namespace,n_classes,covered_cells,unassigned_cells,membership_rows,overlap,complete_partition,scope_note_cn
+known_fine_anatomy,12,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+official_annotation_raw,11,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+legacy52,52,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+evidence_state,4,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+stable8,8,965,4212,965,False,False,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+consensus9,9,1015,4162,1015,False,False,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+scale_units40,40,3848,1329,5431,True,False,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+finest20,20,2125,3052,2125,False,False,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+original61,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+global_full_K20_s2026110301,20,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_full_K40_s2026110301,40,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_full_K80_s2026110301,80,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_full_K80_s2026110302,80,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_full_K80_s2026110303,80,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_full_K80_s2026110304,80,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_full_K80_s2026110305,80,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_full_K80_s2026110306,80,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_full_K160_s2026110301,160,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_full_width20_K80_s2026110301,80,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_full_width80_K80_s2026110301,80,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_identity_position_K20_s2026110301,20,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_identity_position_K40_s2026110301,40,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_identity_position_K80_s2026110301,80,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_identity_position_K160_s2026110301,160,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_partner_identity_K20_s2026110301,20,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_partner_identity_K40_s2026110301,40,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_partner_identity_K80_s2026110301,80,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_partner_identity_K160_s2026110301,160,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_soma_tree_K20_s2026110301,20,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_soma_tree_K40_s2026110301,40,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_soma_tree_K80_s2026110301,80,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+global_soma_tree_K160_s2026110301,160,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+parent_refinement_full,219,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+parent_refinement_identity_position,219,5177,0,5177,False,True,见���协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+parent_refinement_soma_tree,219,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+focal61_full_s2026110310,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_full_s2026110311,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_full_s2026110312,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_full_s2026110313,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_full_s2026110314,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_full_s2026110315,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_full_s2026110316,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_full_s2026110317,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_full_s2026110318,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_full_s2026110319,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_identity_position_s2026110310,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_identity_position_s2026110311,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_identity_position_s2026110312,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_identity_position_s2026110313,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_identity_position_s2026110314,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_identity_position_s2026110315,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_identity_position_s2026110316,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_identity_position_s2026110317,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_identity_position_s2026110318,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_identity_position_s2026110319,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_soma_tree_s2026110310,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_soma_tree_s2026110311,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_soma_tree_s2026110312,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_soma_tree_s2026110313,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_soma_tree_s2026110314,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_soma_tree_s2026110315,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_soma_tree_s2026110316,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_soma_tree_s2026110317,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_soma_tree_s2026110318,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+focal61_soma_tree_s2026110319,3,61,5116,61,False,False,仅原61细胞；其余细胞在该分析范围之外
+historical_wu_20261026_density_refinement,252,5177,0,9468,True,False,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+historical_wu_20261027_branch_contact_model,116,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+historical_wu_20261027_local_tree_refinement,360,5177,0,15531,True,False,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+historical_wu_20261028_open_tools,30,5177,0,5177,False,True,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+historical_wu_20261028_parent_free_model,172,5177,0,20708,True,False,见原协议；covered_cells包括明确的残余角色，未覆盖不表示没有生物功能
+historical_spatial_explanation_registry,422,61,5116,7796,True,False,仅原61细胞；其余细胞在该分析范围之外
+historical_spatial_explanation_resolution_solver_audit,20,61,5116,305,True,False,仅原61细胞；其余细胞在该分析范围之外

wu_20261109_full_atlas_briefing/data/numerical_grid_summary.csv

@@ -0,0 +1,6 @@
+condition,total_count,responding_cells,count_cells_changed_vs_original,count_cells_changed_vs_sample_control,maxabs_onset_vs_original_ms,maxabs_onset_vs_sample_ms,max_soma_RMS_vs_sample_mV,KCmesh_um,MBONmesh_um,dt_ms,initiation_sample_ms
+async_R1_m4,188,32,0,0,0.0,0.0875000000000341,0.0,0.25,0.5,0.0025,0.025
+async_R1_L_sample_control,188,32,0,0,0.0875000000000341,0.0,0.0,0.25,0.5,0.0025,0.0125
+async_R1_L_time_only,192,32,3,3,3.6374999999999886,3.6374999999999886,12.730232467094527,0.25,0.5,0.00125,0.0125
+async_R1_L_space_only,158,32,30,30,2.474999999999966,2.474999999999966,13.31578715421701,0.125,0.25,0.0025,0.0125
+async_R1_L_joint,125,32,32,32,5.149999999999977,5.149999999999977,13.207272204971794,0.125,0.25,0.00125,0.0125

wu_20261109_full_atlas_briefing/data/profile_cells.csv

@@ -0,0 +1,65 @@
+weight_nS,gid,root_id,eligible,amplitude_mV,fine_amplitude_mV,observed_refinement_rms_mV,orientation_refinement_rms,orientation_resolved
+0.1,0,720575940606579465,False,0.0,0.0,0.0,,False
+0.1,1,720575940606698697,False,0.0,0.0,0.0,,False
+0.1,2,720575940615832577,True,0.05519420266499547,0.05520126857541504,3.746872405832935e-05,0.0006666295695448744,True
+0.1,3,720575940618341952,True,0.029707565305681492,0.02973465136844693,6.469999362036444e-05,0.001976959446005633,True
+0.1,4,720575940618662939,False,0.0,0.0,0.0,,False
+0.1,5,720575940619077217,False,0.0,0.0,0.0,,False
+0.1,6,720575940619897734,False,0.0,0.0,0.0,,False
+0.1,7,720575940620333958,False,0.0,0.0,0.0,,False
+0.1,8,720575940621606755,False,0.0,0.0,0.0,,False
+0.1,9,720575940621770010,True,0.014868670072770734,0.014863661504821163,1.3058765756759203e-05,0.0008112437116015559,True
+0.1,10,720575940622003316,False,0.0,0.0,0.0,,False
+0.1,11,720575940622412838,False,0.0,0.0,0.0,,False
+0.1,12,720575940622836455,True,0.006864764352826571,0.006855454764740353,2.355516853051703e-05,0.0031540918288755883,True
+0.1,13,720575940623973961,False,0.0,0.0,0.0,,False
+0.1,14,720575940624107755,False,0.0,0.0,0.0,,False
+0.1,15,720575940625328915,False,0.0,0.0,0.0,,False
+0.1,16,720575940625500304,False,0.0,0.0,0.0,,False
+0.1,17,720575940626305028,True,0.009855169199871075,0.009834756578099344,4.605182361408828e-05,0.004193079467061918,True
+0.1,18,720575940627585028,False,0.0,0.0,0.0,,False
+0.1,19,720575940627626244,False,0.0,0.0,0.0,,False
+0.1,20,720575940629319106,True,0.017127577839971285,0.01711212209070044,3.5440801479330636e-05,0.001862931598570388,True
+0.1,21,720575940629507706,True,0.01689198813838252,0.016904243719487587,4.754411328003443e-05,0.0027184917708093924,True
+0.1,22,720575940630331244,True,0.015776790577302226,0.01578144977496878,2.4837775548420877e-05,0.001546148578499754,True
+0.1,23,720575940630775891,False,0.0,0.0,0.0,,False
+0.1,24,720575940630923693,False,0.0,0.0,0.0,,False
+0.1,25,720575940632437331,True,0.0029111109740949507,0.0029083956493533387,7.721167592222387e-06,0.0024840467276466016,True
+0.1,26,720575940633103059,False,0.0,0.0,0.0,,False
+0.1,27,720575940634514533,False,0.0,0.0,0.0,,False
+0.1,28,720575940634664674,False,0.0,0.0,0.0,,False
+0.1,29,720575940643769544,False,0.0,0.0,0.0,,False
+0.1,30,720575940644413732,False,0.0,0.0,0.0,,False
+0.1,31,720575940649022841,True,0.39244155820806687,0.3923408536369318,0.00016909449628021894,0.00034617605483753497,True
+1.0,0,720575940606579465,False,9.455700418835812,9.456692318816552,0.2112224834704528,0.022336692950366843,True
+1.0,1,720575940606698697,False,12.52242877366163,12.524712614747973,0.4173731710666863,0.03332651161756826,True
+1.0,2,720575940615832577,True,8.797536316055393,8.791164850933061,0.099979227076105,0.011345466173425517,True
+1.0,3,720575940618341952,True,9.258017460092212,9.252125441778011,0.10298138398544018,0.011108795585248075,True
+1.0,4,720575940618662939,False,10.123013226416738,10.12306889098427,0.21663105474929004,0.02139979936288592,True
+1.0,5,720575940619077217,False,10.06395310112888,10.056182870595384,0.1862179872405347,0.01849448913969597,True
+1.0,6,720575940619897734,False,10.015603441870631,10.014690799999526,0.25325266269705177,0.025286799630732177,True
+1.0,7,720575940620333958,False,11.52440885979233,11.52574087106008,0.23252633618738044,0.020175358866777092,True
+1.0,8,720575940621606755,False,9.665302284727106,9.66681616666674,0.2874792590810535,0.029740690298784872,True
+1.0,9,720575940621770010,True,11.276877111252,11.273537567549653,0.41266611049054897,0.03659822738237355,True
+1.0,10,720575940622003316,False,10.104407264969268,10.112102000893682,0.25373892807577225,0.02509060721422505,True
+1.0,11,720575940622412838,False,12.640639744561721,12.63916658444991,0.4128955196445501,0.03266582717739691,True
+1.0,12,720575940622836455,True,9.560631976406425,9.55975074006448,0.2097040158637712,0.02193493379124783,True
+1.0,13,720575940623973961,False,9.990078246968952,9.987520658323298,0.22683009833283566,0.022707001172217366,True
+1.0,14,720575940624107755,False,11.502321440032176,11.499678331049033,0.23425462638421554,0.020366900148571312,True
+1.0,15,720575940625328915,False,11.113802603282398,11.11421921629869,0.2638869111523419,0.023743595761715353,True
+1.0,16,720575940625500304,False,10.202480017048567,10.19761825162353,0.198754737695885,0.019479834821072465,True
+1.0,17,720575940626305028,True,10.561249805061168,10.564147492359966,0.20019763047525607,0.01895128008100888,True
+1.0,18,720575940627585028,False,10.283916550157183,10.285718276529032,0.29371131851960136,0.028557219986060162,True
+1.0,19,720575940627626244,False,9.702754298012382,9.701083795298585,0.20201854168141306,0.020821822425021758,True
+1.0,20,720575940629319106,True,9.207047640504712,9.202213898829935,0.22624178158058866,0.024573516502659373,True
+1.0,21,720575940629507706,True,10.00509690486616,9.999635181664813,0.1053653889172509,0.010519885010080136,True
+1.0,22,720575940630331244,True,8.757236279048495,8.757149511249834,0.23889143261453727,0.027279451271688276,True
+1.0,23,720575940630775891,False,10.464598610949741,10.467288258471577,0.28755161649792293,0.027473781166952604,True
+1.0,24,720575940630923693,False,9.41932935089379,9.420992082749892,0.22148823624332087,0.023511488024251247,True
+1.0,25,720575940632437331,True,9.822315844446946,9.82086793684729,0.2666155275606638,0.027145456627820082,True
+1.0,26,720575940633103059,False,11.780726745487145,11.780306649710175,0.4018849286865486,0.03411435334295469,True
+1.0,27,720575940634514533,False,10.393850472524734,10.400685656156929,0.22502385652920542,0.021632608941537793,True
+1.0,28,720575940634664674,False,11.50366064983821,11.50157360363157,0.45789026684742323,0.03980707236656624,True
+1.0,29,720575940643769544,False,12.866647036702991,12.861177375454735,0.18343145137355696,0.014253042600166655,True
+1.0,30,720575940644413732,False,9.7072440777819,9.708127369552347,0.22032112478037982,0.022695353008483427,True
+1.0,31,720575940649022841,True,8.013528675231395,8.007661212521947,0.11148490059539114,0.013897893872408801,True

wu_20261109_full_atlas_briefing/data/schedule_masks.csv

@@ -0,0 +1,21 @@
+schedule,mask,total_waveforms,responding_cells,mean_soma_mV
+synchronous,1,559,32,-51.889245951097735
+synchronous,2,0,0,-69.4666423914508
+synchronous,4,0,0,-69.41814503150141
+synchronous,7,559,32,-51.8641808874644
+jittered_seed20261106,1,524,32,-52.760763274815346
+jittered_seed20261106,2,0,0,-69.46847808652672
+jittered_seed20261106,4,0,0,-69.42002517583455
+jittered_seed20261106,7,551,32,-51.9487440983519
+jittered_seed20261107,1,406,32,-56.14146997407278
+jittered_seed20261107,2,0,0,-69.47332323800387
+jittered_seed20261107,4,0,0,-69.424783566559
+jittered_seed20261107,7,435,32,-55.183354635123095
+jittered_seed20261108,1,494,32,-53.45230047240095
+jittered_seed20261108,2,0,0,-69.4695569787789
+jittered_seed20261108,4,0,0,-69.42110327934355
+jittered_seed20261108,7,465,32,-54.42511696460077
+bursty_three_epochs,1,96,32,-67.22704586167686
+bursty_three_epochs,2,96,32,-67.27364448303024
+bursty_three_epochs,4,96,32,-67.2604111604264
+bursty_three_epochs,7,96,32,-67.20574192974635

wu_20261109_full_atlas_briefing/data/source_original61_with_simulation_order.csv

@@ -0,0 +1,62 @@
+root_id,candidate,cluster_size,cohort_order,group_order,official_KC_subtype,hemisphere,soma_x_um,soma_y_um,soma_z_um,dataset_version,coordinate_space,simulation_gid,in_current_32_KC_simulation
+720575940606579465,abp_R_K3_C0,32,1,1,KCab-p,right,610.624,104.0,171.96,FlyWire v783,FAFB14 um,0,True
+720575940606698697,abp_R_K3_C0,32,2,2,KCab-p,right,592.256,105.152,168.08,FlyWire v783,FAFB14 um,1,True
+720575940615832577,abp_R_K3_C0,32,3,3,KCab-p,right,617.44,106.08,169.04,FlyWire v783,FAFB14 um,2,True
+720575940618341952,abp_R_K3_C0,32,4,4,KCab-p,right,603.392,104.48,169.32,FlyWire v783,FAFB14 um,3,True
+720575940618662939,abp_R_K3_C0,32,5,5,KCab-p,right,577.28,112.352,164.72,FlyWire v783,FAFB14 um,4,True
+720575940619077217,abp_R_K3_C0,32,6,6,KCab-p,right,624.704,118.872,191.68,FlyWire v783,FAFB14 um,5,True
+720575940619897734,abp_R_K3_C0,32,7,7,KCab-p,right,622.656,104.992,175.84,FlyWire v783,FAFB14 um,6,True
+720575940620333958,abp_R_K3_C0,32,8,8,KCab-p,right,605.28,117.76,165.08,FlyWire v783,FAFB14 um,7,True
+720575940621606755,abp_R_K3_C0,32,9,9,KCab-p,right,619.072,104.352,167.04,FlyWire v783,FAFB14 um,8,True
+720575940621770010,abp_R_K3_C0,32,10,10,KCab-p,right,611.424,102.496,168.76,FlyWire v783,FAFB14 um,9,True
+720575940622003316,abp_R_K3_C0,32,11,11,KCab-p,right,616.736,108.768,177.0,FlyWire v783,FAFB14 um,10,True
+720575940622412838,abp_R_K3_C0,32,12,12,KCab-p,right,604.0,111.072,169.92,FlyWire v783,FAFB14 um,11,True
+720575940622836455,abp_R_K3_C0,32,13,13,KCab-p,right,619.744,105.6,172.64,FlyWire v783,FAFB14 um,12,True
+720575940623973961,abp_R_K3_C0,32,14,14,KCab-p,right,616.288,106.72,174.2,FlyWire v783,FAFB14 um,13,True
+720575940624107755,abp_R_K3_C0,32,15,15,KCab-p,right,597.536,103.392,166.16,FlyWire v783,FAFB14 um,14,True
+720575940625328915,abp_R_K3_C0,32,16,16,KCab-p,right,608.064,121.024,187.8,FlyWire v783,FAFB14 um,15,True
+720575940625500304,abp_R_K3_C0,32,17,17,KCab-p,right,620.104,111.376,175.72,FlyWire v783,FAFB14 um,16,True
+720575940626305028,abp_R_K3_C0,32,18,18,KCab-p,right,590.016,103.296,163.96,FlyWire v783,FAFB14 um,17,True
+720575940627585028,abp_R_K3_C0,32,19,19,KCab-p,right,626.4,106.848,167.84,FlyWire v783,FAFB14 um,18,True
+720575940627626244,abp_R_K3_C0,32,20,20,KCab-p,right,618.944,105.216,176.96,FlyWire v783,FAFB14 um,19,True
+720575940629319106,abp_R_K3_C0,32,21,21,KCab-p,right,604.224,101.152,167.0,FlyWire v783,FAFB14 um,20,True
+720575940629507706,abp_R_K3_C0,32,22,22,KCab-p,right,600.256,100.672,162.04,FlyWire v783,FAFB14 um,21,True
+720575940630331244,abp_R_K3_C0,32,23,23,KCab-p,right,598.304,108.192,168.16,FlyWire v783,FAFB14 um,22,True
+720575940630775891,abp_R_K3_C0,32,24,24,KCab-p,right,596.64,100.032,160.2,FlyWire v783,FAFB14 um,23,True
+720575940630923693,abp_R_K3_C0,32,25,25,KCab-p,right,609.856,116.96,188.24,FlyWire v783,FAFB14 um,24,True
+720575940632437331,abp_R_K3_C0,32,26,26,KCab-p,right,599.488,108.736,157.52,FlyWire v783,FAFB14 um,25,True
+720575940633103059,abp_R_K3_C0,32,27,27,KCab-p,right,601.28,124.96,167.16,FlyWire v783,FAFB14 um,26,True
+720575940634514533,abp_R_K3_C0,32,28,28,KCab-p,right,617.632,101.216,169.68,FlyWire v783,FAFB14 um,27,True
+720575940634664674,abp_R_K3_C0,32,29,29,KCab-p,right,611.008,107.488,179.16,FlyWire v783,FAFB14 um,28,True
+720575940643769544,abp_R_K3_C0,32,30,30,KCab-p,right,600.864,105.952,166.76,FlyWire v783,FAFB14 um,29,True
+720575940644413732,abp_R_K3_C0,32,31,31,KCab-p,right,590.72,106.784,169.64,FlyWire v783,FAFB14 um,30,True
+720575940649022841,abp_R_K3_C0,32,32,32,KCab-p,right,606.336,110.816,173.72,FlyWire v783,FAFB14 um,31,True
+720575940607054770,abp_R_K3_C1,15,33,1,KCab-p,right,638.496,175.84,213.24,FlyWire v783,FAFB14 um,,False
+720575940609371086,abp_R_K3_C1,15,34,2,KCab-p,right,653.152,172.64,207.0,FlyWire v783,FAFB14 um,,False
+720575940610823010,abp_R_K3_C1,15,35,3,KCab-p,right,632.992,179.008,214.4,FlyWire v783,FAFB14 um,,False
+720575940612767126,abp_R_K3_C1,15,36,4,KCab-p,right,617.92,168.192,212.2,FlyWire v783,FAFB14 um,,False
+720575940615671494,abp_R_K3_C1,15,37,5,KCab-p,right,632.896,182.048,214.2,FlyWire v783,FAFB14 um,,False
+720575940617478427,abp_R_K3_C1,15,38,6,KCab-p,right,621.76,173.76,209.08,FlyWire v783,FAFB14 um,,False
+720575940617771521,abp_R_K3_C1,15,39,7,KCab-p,right,626.688,169.696,213.36,FlyWire v783,FAFB14 um,,False
+720575940621925417,abp_R_K3_C1,15,40,8,KCab-p,right,624.032,156.544,208.96,FlyWire v783,FAFB14 um,,False
+720575940627085578,abp_R_K3_C1,15,41,9,KCab-p,right,636.608,179.296,213.32,FlyWire v783,FAFB14 um,,False
+720575940628045318,abp_R_K3_C1,15,42,10,KCab-p,right,619.008,165.504,209.92,FlyWire v783,FAFB14 um,,False
+720575940628826873,abp_R_K3_C1,15,43,11,KCab-p,right,641.856,177.696,212.12,FlyWire v783,FAFB14 um,,False
+720575940631485251,abp_R_K3_C1,15,44,12,KCab-p,right,641.696,174.016,212.12,FlyWire v783,FAFB14 um,,False
+720575940632366623,abp_R_K3_C1,15,45,13,KCab-p,right,636.352,175.072,210.2,FlyWire v783,FAFB14 um,,False
+720575940633226603,abp_R_K3_C1,15,46,14,KCab-p,right,623.968,162.496,209.96,FlyWire v783,FAFB14 um,,False
+720575940634351200,abp_R_K3_C1,15,47,15,KCab-p,right,626.848,173.184,214.4,FlyWire v783,FAFB14 um,,False
+720575940608191836,abp_R_K3_C2,14,48,1,KCab-p,right,681.664,112.256,164.56,FlyWire v783,FAFB14 um,,False
+720575940610580377,abp_R_K3_C2,14,49,2,KCab-p,right,686.304,126.016,179.12,FlyWire v783,FAFB14 um,,False
+720575940615196063,abp_R_K3_C2,14,50,3,KCab-p,right,675.692,124.728,183.04,FlyWire v783,FAFB14 um,,False
+720575940616860318,abp_R_K3_C2,14,51,4,KCab-p,right,670.632,120.672,182.36,FlyWire v783,FAFB14 um,,False
+720575940617750173,abp_R_K3_C2,14,52,5,KCab-p,right,678.752,118.976,172.16,FlyWire v783,FAFB14 um,,False
+720575940619684688,abp_R_K3_C2,14,53,6,KCab-p,right,677.004,127.976,185.08,FlyWire v783,FAFB14 um,,False
+720575940619937718,abp_R_K3_C2,14,54,7,KCab-p,right,672.96,130.88,185.32,FlyWire v783,FAFB14 um,,False
+720575940620700769,abp_R_K3_C2,14,55,8,KCab-p,right,696.112,122.052,166.44,FlyWire v783,FAFB14 um,,False
+720575940621762330,abp_R_K3_C2,14,56,9,KCab-p,right,689.152,126.592,169.88,FlyWire v783,FAFB14 um,,False
+720575940625126448,abp_R_K3_C2,14,57,10,KCab-p,right,679.524,123.96,182.44,FlyWire v783,FAFB14 um,,False
+720575940626043454,abp_R_K3_C2,14,58,11,KCab-p,right,698.432,130.912,176.24,FlyWire v783,FAFB14 um,,False
+720575940628514156,abp_R_K3_C2,14,59,12,KCab-p,right,628.672,103.68,169.44,FlyWire v783,FAFB14 um,,False
+720575940630949187,abp_R_K3_C2,14,60,13,KCab-p,right,659.296,115.52,182.64,FlyWire v783,FAFB14 um,,False
+720575940634969701,abp_R_K3_C2,14,61,14,KCab-p,right,664.8,114.24,177.16,FlyWire v783,FAFB14 um,,False