Archive 4 checkpoints (1-4)

paper-releases/pbisc-diffusion/updates/20261003T174003Z-scsemi-readout-sensitivity/diffusion/dataset/server_1/model8/plans/paired_finetune_intradataset_20260927/scores/giroux2022/fewshot_n5/draw1_scsemi5rt/README.md

@@ -0,0 +1,73 @@
+# score_ft.py output: giroux2022, 10 scored donors
+
+Arms: scsemi5rt.  Every arm is scored on the same donors against the same observed reference.
+
+## Inputs and fixed contracts
+- Observed reference: `/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/scoring_cache/giroux2022/reference.npz` (sha dd0152f7bde4), built from datadir:/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/data/giroux2022;
+  observed cells sum exactly to the features pseudobulk of every donor. Observed labels: /home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/scoring_cache/giroux2022/observed_labels/labels/labels.parquet.
+- Generated cells: `generated_counts` read out once with Poisson(seed 20260916) over the **whole NPZ in row order**
+  (`poisson_readout.poisson_integerise`), then the scored donors' rows. These integer counts are used for labels, embedding,
+  depth and the contrast. Pseudobulk uses the summed `generated_counts` (rates), as `score_generated.py`.
+- Labels: frozen f01common scANVI (`predict_scanvi.py`, CPU, zero query epochs), six-type gate ['T cell', 'B cell', 'Monocyte', 'NK cell', 'DC', 'Other']; query genes =
+  the 1,006-gene annotation list minus genes the dataset's sc assay did not measure in every sample (1006 genes).
+- Scoring genes: 6k panel genes measured by both assays in every sample of the dataset (5563 genes).
+
+## Definitions
+1. **Composition**: fraction of each donor's cells per six-type label (all generated cells; all observed QC cells).
+   `comp_l1` = sum over the six labels of |generated - observed| (immicro_common.l1). `comp_l1_obs_loo_null` = L1 of the
+   donor's observed composition to the mean of the *other scored donors'* observed compositions. `composition_top1` =
+   fraction of donors whose generated composition is closest (L1) to their own observed composition.
+2. **Pseudobulk**: `pb_r` = Pearson r of CPM-log1p(summed generated_counts) vs CPM-log1p(observed pseudobulk) on the scoring
+   genes (evaluate_realbulk.cpm_log1p / paired_r). `pb_top1` = the donor's generated pseudobulk is closest (Pearson) to its own
+   observed pseudobulk among the 10 scored donors (immicro_common.retrieval = evaluate_realbulk.top1 arithmetic).
+3. **Precision / coverage / density** (stage_b_embed.py functions): per-cell CP10K over the scoring genes + log1p; PCA-50
+   (randomized, random_state 20260925) and k = 10 real-real k-NN balls fitted on **R1 of the scored donors only**.
+   Reference rule (independent of fold and arm): per donor, cells in canonical order (library_id+barcode), one
+   permutation from `default_rng([20260925, uint64(sha256(donor)[:8])])`; R1 = first n, R2 = next n,
+   n = min(150, n_cells // 2) (1500 R1 cells here). Generated query: per donor n cells (the donor's R1 size),
+   without replacement, `default_rng([20260926, same key])`. Precision = fraction of generated cells inside any R1 ball
+   (Kynkaanniemi 2019); coverage = fraction of R1 cells whose ball holds a generated cell; density = Naeem 2020.
+   Per-donor precision = that donor's generated cells; per-donor coverage = that donor's R1 cells (the pooled values are
+   their n-weighted means; plain means when every donor has 150 reference cells). R2 (held-out real cells of the same donors) is the
+   perfect-generator control: precision 1.0000, coverage 1.0000.
+   **Depth matched**: stage_b_embed depth matching run pairwise (R1 vs this arm): per donor and within-donor rank of detected
+   genes, both cells are thinned by UMI sampling without replacement to the smaller detected-gene count; PCA kept from
+   un-thinned R1, radii recomputed on thinned R1; fresh rng(20260929) per arm. R2 control: precision
+   0.9993, coverage 1.0000.
+4. **Monocyte %** = 100 x Monocyte fraction (all cells) per donor; **median UMI per cell** = median over cells of the
+   Poisson-read (generated) / observed counts summed over the scoring genes (per donor, and pooled).
+5. **Group contrast** (T, B, Monocyte, NK): case groups ['severe'] (n = 6) vs control groups ['mild'] (n = 4); rule: explicit --case-groups / --control-groups; groups from features groups.
+   Genes per type = scoring genes detected (count > 0) in >= 1 % of the scored donors' observed
+   cells of that type. Primary statistic `mean_cp10k`: per-cell CP10K over the scoring genes; per donor the mean over its
+   cells of the type (donors with >= 1 such cell); group mean = mean over the group's donors (donor-balanced);
+   logFC = log2(case mean + 1) - log2(control mean + 1). Secondary `pseudobulk_log1p_cpm`: the SLE benchmark evaluator's
+   statistic (donor x type summed counts, log1p CPM over the scoring genes, mean case - mean control;
+   evaluate_generated_vs_truth.group_log_fold_change). Both are computed on the generated cells (their frozen labels) and on
+   the observed cells; `r_generated` = Pearson r of the two logFC vectors. Broadcast baseline `r_broadcast`: the same
+   observed vector against the logFC of the input bulk (condition NPZ `bulk_counts` of the arm, scoring genes, per-donor
+   CP10K resp. log1p CPM, group means). `partial_r_generated_given_broadcast`: r of the OLS residuals of observed and
+   generated logFC on the bulk logFC (residual_specificity.py).
+6. **Top-50 DEG overlap** (contrast.csv rows `statistic = welch_top50_deg`; summary `deg_top50`): the SLE benchmark's score (`total_top50_overlap` of
+   `bulk2sc-paper/scripts/evaluate_generated_vs_truth.py`, Welch backend: the "top-50 DEG 81 / 84" of the final model), run
+   with the evaluator's own functions on the same donor x type pseudobulks as the contrast. Per cell type, observed and
+   generated separately: donor x type summed counts on all 5563 scoring genes (donors of the two groups with
+   >= 1 cell of the type) -> `welch_bh_deg` (log1p CPM, two-sided Welch t-test case vs control, Benjamini-Hochberg over the
+   genes with a defined p-value; needs >= 2 donors per group) -> `top_deg_genes` (FDR < 0.05, descending |logFC|,
+   first 50; shorter when fewer genes pass). `top50_overlap` = |observed list & generated list|; `top50_jaccard`
+   the evaluator's Jaccard. Summary `sum` = total over the scored types (the quantity reported as 81), `max` = the sum of
+   the observed list lengths (the "of 262"), `sum_4type` = T + B + Monocyte + NK. A type is scored when its observed test
+   is defined; an arm whose generated test is undefined for a scored type gets overlap 0. Types: T cell, B cell, Monocyte, NK cell, DC
+   (aggregates: /home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/scoring_cache/giroux2022/deg_reference.npz).
+   **Broadcast analogue** (`top50_overlap_broadcast`, as `sle_broadcast.py` `broadcast_input_bulk`; not part of the
+   published benchmark): the same test on the arm's condition bulk_counts (scoring genes, the 6 + 4
+   donors), one whole-PBMC top-50 list, compared with every type's observed list.
+
+## Files
+- `per_donor.csv`: one row per arm x donor (all per-donor quantities above).
+- `summary.json`: per arm the donor means / pooled values, provenance (NPZ sha, receipt check, generation metadata,
+  Poisson nnz of the whole NPZ, annotation contract), and the reference block (PCA, R2 control).
+- `contrast.csv`: per arm x cell type x statistic (`mean_cp10k`, `pseudobulk_log1p_cpm`: logFC r; `welch_top50_deg`:
+  top-50 DEG overlap, T / B / Monocyte / NK / DC; columns of the other kind are empty).
+- `arms/<arm>/labels/` (predict_scanvi.py labels + manifest), `arms/<arm>/generated_cells.parquet` (NPZ row, donor, label,
+  UMI on the scoring genes, P/C subsample flag), `arms/<arm>/deg_top50_lists.csv` (observed, generated and broadcast
+  top-50 gene lists, Ensembl ids, rank order).

paper-releases/pbisc-diffusion/updates/20261003T174003Z-scsemi-readout-sensitivity/diffusion/dataset/server_1/model8/plans/paired_finetune_intradataset_20260927/scores/giroux2022/fewshot_n5/draw1_scsemi5rt/contrast.csv

@@ -0,0 +1,14 @@
+arm,celltype,statistic,n_genes,r_generated,r_broadcast,partial_r_generated_given_broadcast,r_generated_vs_broadcast,n_case_donors_obs,n_control_donors_obs,n_case_donors_gen,n_control_donors_gen,n_cells_obs,n_cells_gen,deg_scored,top50_overlap,top50_jaccard,obs_top_list_size,gen_top_list_size,n_sig_obs,n_sig_gen,top50_overlap_broadcast,top50_jaccard_broadcast,broadcast_top_list_size,n_sig_broadcast
+scsemi5rt,T cell,mean_cp10k,2938,0.461007373754063,0.6850650089880258,0.20666962284446275,0.48016416220888797,6,4,6,4,58362,7581,,,,,,,,,,,
+scsemi5rt,T cell,pseudobulk_log1p_cpm,2938,0.372771428325728,0.778305137206687,0.07725352233801083,0.42246438353816773,6,4,6,4,58362,7581,,,,,,,,,,,
+scsemi5rt,B cell,mean_cp10k,3121,0.3267962363095575,0.43529553800990617,0.22592752282280693,0.30587248786425675,6,4,6,4,6177,827,,,,,,,,,,,
+scsemi5rt,B cell,pseudobulk_log1p_cpm,3121,0.14450465993784575,0.5433173120170087,0.01790483810602624,0.23910348470320145,6,4,6,4,6177,827,,,,,,,,,,,
+scsemi5rt,Monocyte,mean_cp10k,3066,0.4908860569310003,0.4929362363245346,0.3920381374930296,0.3468159100696941,6,4,6,4,18467,4261,,,,,,,,,,,
+scsemi5rt,Monocyte,pseudobulk_log1p_cpm,3066,0.2155998404805981,0.6439809335079176,0.11921659066254596,0.19590885839874383,6,4,6,4,18467,4261,,,,,,,,,,,
+scsemi5rt,NK cell,mean_cp10k,3018,0.4075719413194648,0.5413246923269038,0.22159569045728014,0.4446136040316459,6,4,6,4,9050,1887,,,,,,,,,,,
+scsemi5rt,NK cell,pseudobulk_log1p_cpm,3018,0.2565887762990696,0.6587138726193354,0.03486750803315499,0.35225643617782826,6,4,6,4,9050,1887,,,,,,,,,,,
+scsemi5rt,T cell,welch_top50_deg,5563,,,,,6,4,6,4,58362,7581,True,0,0.0,50,0,99,0,6,0.10526315789473684,13,13
+scsemi5rt,B cell,welch_top50_deg,5563,,,,,6,4,6,4,6177,827,True,0,0.0,4,0,4,0,4,0.3076923076923077,13,13
+scsemi5rt,Monocyte,welch_top50_deg,5563,,,,,6,4,6,4,18467,4261,True,0,0.0,5,0,5,0,5,0.38461538461538464,13,13
+scsemi5rt,NK cell,welch_top50_deg,5563,,,,,6,4,6,4,9050,1887,True,0,0.0,5,0,5,0,5,0.38461538461538464,13,13
+scsemi5rt,DC,welch_top50_deg,5563,,,,,6,4,5,4,1271,81,True,0,0.0,4,1,4,1,4,0.3076923076923077,13,13

paper-releases/pbisc-diffusion/updates/20261003T174003Z-scsemi-readout-sensitivity/diffusion/dataset/server_1/model8/plans/paired_finetune_intradataset_20260927/scores/giroux2022/fewshot_n5/draw1_scsemi5rt/per_donor.csv

@@ -0,0 +1,11 @@
+arm,donor,group,n_generated_cells,n_observed_cells,n_reference_cells,gen_T cell,obs_T cell,gen_B cell,obs_B cell,gen_Monocyte,obs_Monocyte,gen_NK cell,obs_NK cell,gen_DC,obs_DC,gen_Other,obs_Other,comp_l1,comp_l1_obs_loo_null,pb_r,pb_own_rank,pb_top1,precision,coverage,precision_depth_matched,coverage_depth_matched,mono_pct_gen,mono_pct_obs,median_umi_gen,median_umi_obs
+scsemi5rt,450905_d0,mild,1500,5914,150,0.22,0.30943523841731485,0.02266666666666667,0.036185322962461956,0.5906666666666667,0.4293202570172472,0.15933333333333333,0.12089956036523504,0.002,0.009976327358809604,0.005333333333333333,0.09418329387893135,0.3995603652350355,0.6721216782161549,0.6208232172357986,1,True,0.9866666666666667,0.6266666666666667,0.5133333333333333,0.4066666666666667,59.06666666666667,42.93202570172472,465.0,574.5
+scsemi5rt,450905_d14,mild,1500,7407,150,0.49133333333333334,0.4422843256379101,0.034,0.02929661131362225,0.23666666666666666,0.27420008100445525,0.20866666666666667,0.15782368030241664,0.009333333333333334,0.02983664101525584,0.02,0.06655866072633995,0.20919076549210208,0.4172767195297901,0.5857371346654185,1,True,0.9933333333333333,0.5466666666666666,0.5866666666666667,0.2,23.666666666666668,27.420008100445525,464.0,558.0
+scsemi5rt,0B943B_d0,mild,1500,11344,150,0.49866666666666665,0.6220028208744711,0.027333333333333334,0.08312764456981664,0.2826666666666667,0.12588152327221439,0.17266666666666666,0.11865303244005641,0.007333333333333333,0.023977433004231313,0.011333333333333334,0.026357545839210156,0.4215975552421251,0.27288152831947954,0.5940650870242553,7,False,1.0,0.47333333333333333,0.58,0.1,28.26666666666667,12.588152327221438,465.0,539.0
+scsemi5rt,0B943B_d14,mild,1500,15650,150,0.5226666666666666,0.5652396166134185,0.12466666666666666,0.11175718849840256,0.19333333333333333,0.17086261980830672,0.10133333333333333,0.09789137380191694,0.0026666666666666666,0.009968051118210863,0.05533333333333333,0.04428115015974441,0.09974866879659208,0.1514797576931834,0.6576400440226762,3,False,0.9866666666666667,0.7266666666666667,0.4066666666666667,0.26,19.333333333333332,17.08626198083067,464.0,461.0
+scsemi5rt,180E1A_d0,severe,1500,7157,150,0.622,0.7019700992035769,0.09466666666666666,0.04135811094033813,0.16133333333333333,0.13175911694844208,0.10066666666666667,0.0655302501047925,0.004666666666666667,0.013273718038284197,0.016666666666666666,0.04610870476456616,0.23603837734618785,0.37947485892421606,0.6102633842128057,7,False,0.9933333333333333,0.6066666666666667,0.4266666666666667,0.21333333333333335,16.133333333333333,13.175911694844208,465.0,433.0
+scsemi5rt,180E1A_d7,severe,1500,10591,150,0.538,0.7348692285903126,0.03266666666666666,0.04909829100179398,0.212,0.09064299877254273,0.18533333333333332,0.08280615617033331,0.009333333333333334,0.010480596733075253,0.02266666666666667,0.03210272873194221,0.4477683567809146,0.4452731176976873,0.609295203012962,4,False,1.0,0.6066666666666667,0.44,0.11333333333333333,21.2,9.064299877254273,466.0,408.0
+scsemi5rt,180E1A_d14,severe,1500,12511,150,0.6633333333333333,0.680840860043162,0.10066666666666667,0.05147470226200943,0.122,0.12277196067460634,0.07933333333333334,0.0800095915594277,0.007333333333333333,0.009831348413396212,0.027333333333333334,0.05507153704739829,0.09838392880931446,0.33721638060338627,0.6695467253811618,1,True,1.0,0.58,0.52,0.22,12.2,12.277196067460634,464.0,342.0
+scsemi5rt,0BF51C_d0,severe,1500,6510,150,0.21733333333333332,0.4012288786482335,0.020666666666666667,0.01674347158218126,0.596,0.4609831029185868,0.162,0.07726574500768049,0.0,0.006144393241167435,0.004,0.03763440860215054,0.44734869431643626,0.5375551894092265,0.5946427463051597,1,True,0.9866666666666667,0.3933333333333333,0.48,0.34,59.599999999999994,46.09831029185868,465.0,628.0
+scsemi5rt,0BF51C_d7,severe,1500,10105,150,0.608,0.612370113805047,0.052,0.05700148441365661,0.24333333333333335,0.1607125185551707,0.04466666666666667,0.05294408708560119,0.0033333333333333335,0.0073231073725878275,0.048666666666666664,0.10964868876793667,0.1652416295563253,0.18080737356645343,0.5728377786095519,3,False,0.9933333333333333,0.6066666666666667,0.6533333333333333,0.18666666666666668,24.333333333333336,16.071251855517072,465.0,509.0
+scsemi5rt,0BF51C_d14,severe,1500,11419,150,0.6726666666666666,0.6330676941938874,0.042,0.07960416849111131,0.20266666666666666,0.151589456169542,0.044,0.07907872843506436,0.008,0.010508801120938787,0.030666666666666665,0.04615115158945617,0.18135236593980786,0.15136085081896566,0.6232188796271123,1,True,0.9866666666666667,0.5666666666666667,0.5733333333333334,0.24666666666666667,20.266666666666666,15.1589456169542,465.0,541.0

paper-releases/pbisc-diffusion/updates/20261003T174003Z-scsemi-readout-sensitivity/diffusion/dataset/server_1/model8/plans/paired_finetune_intradataset_20260927/scores/giroux2022/fewshot_n5/draw1_scsemi5rt/summary.json

@@ -0,0 +1,475 @@
+{
+ "schema": "paired_ft_scores.v1",
+ "dataset": "giroux2022",
+ "written_utc": "2026-10-03T14:59:42.145655+00:00",
+ "command": [
+  "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/tools/score_ft.py",
+  "--dataset",
+  "giroux2022",
+  "--out",
+  "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/scores/giroux2022/fewshot_n5/draw1_scsemi5rt",
+  "--threads",
+  "6",
+  "--donors",
+  "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/data/giroux2022/fewshot_n5/draw1/test_samples.txt",
+  "--case-groups",
+  "severe",
+  "--control-groups",
+  "mild",
+  "--arm",
+  "scsemi5rt=/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/generated/giroux2022/fewshot_n5/draw1/scsemi5rt__raw.npz"
+ ],
+ "arms": [
+  "scsemi5rt"
+ ],
+ "donors": [
+  "450905_d0",
+  "450905_d14",
+  "0B943B_d0",
+  "0B943B_d14",
+  "180E1A_d0",
+  "180E1A_d7",
+  "180E1A_d14",
+  "0BF51C_d0",
+  "0BF51C_d7",
+  "0BF51C_d14"
+ ],
+ "n_donors": 10,
+ "reference": {
+  "cache": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/scoring_cache/giroux2022",
+  "reference_sha256": "dd0152f7bde41da80e1c71dabf54ee561982f36dc99a336b65fca419880a1593",
+  "features_sha256": "fb41b91617e3ff786d5153c4dc7869d66191ca5947ff28161d24215a24f819e5",
+  "n_r1_cells": 1500,
+  "pca": {
+   "n_components": 50,
+   "explained_variance_ratio_sum": 0.25688886642456055,
+   "n_fit_cells": 1500
+  },
+  "r2_control": {
+   "precision": 1.0,
+   "coverage": 1.0,
+   "density": 6.064666666666667,
+   "n_query": 1500,
+   "depth_matched": {
+    "precision": 0.9993333333333333,
+    "coverage": 1.0,
+    "density": 6.0241999999999996
+   }
+  },
+  "observed_l1_loo_null_mean": 0.3545447454778543
+ },
+ "contrast_definition": {
+  "enabled": true,
+  "rule": "explicit --case-groups / --control-groups",
+  "group_source": "features groups",
+  "case_groups": [
+   "severe"
+  ],
+  "control_groups": [
+   "mild"
+  ],
+  "n_case_donors": 6,
+  "n_control_donors": 4,
+  "n_donors_in_neither": 0,
+  "case": [
+   false,
+   false,
+   false,
+   false,
+   true,
+   true,
+   true,
+   true,
+   true,
+   true
+  ],
+  "ctrl": [
+   true,
+   true,
+   true,
+   true,
+   false,
+   false,
+   false,
+   false,
+   false,
+   false
+  ],
+  "groups": [
+   "mild",
+   "mild",
+   "mild",
+   "mild",
+   "severe",
+   "severe",
+   "severe",
+   "severe",
+   "severe",
+   "severe"
+  ]
+ },
+ "deg_definition": {
+  "enabled": true,
+  "score": "top-50 DEG overlap per cell type (benchmark total_top50_overlap, Welch backend)",
+  "source": "/home/sj_server_1/PBISC/immun-pbmc-fm/bulk2sc-paper/scripts/evaluate_generated_vs_truth.py",
+  "functions": [
+   "welch_bh_deg",
+   "top_deg_genes",
+   "jaccard"
+  ],
+  "reproduces": "utility_studies_final6k_20260925/sle top-50 DEG 81 (1,006 genes) / 84 (2,279 genes) = sum over the six gated types of top50_overlap_count",
+  "top_n": 50,
+  "fdr": 0.05,
+  "n_genes": 5563,
+  "types": [
+   "T cell",
+   "B cell",
+   "Monocyte",
+   "NK cell",
+   "DC"
+  ],
+  "aggregates_source": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/scoring_cache/giroux2022/deg_reference.npz",
+  "aggregates_sha256": "412c44bba7481b0b851c3b12e695f1b7ff662163e1f7849d3a64359dfce6d50e",
+  "aggregates_note": null,
+  "types_not_scored_observed": [],
+  "observed": {
+   "T cell": {
+    "n_significant": 99,
+    "n_tested": 5282,
+    "n_case_donors": 6,
+    "n_control_donors": 4,
+    "n_cells": 58362,
+    "top_list_size": 50
+   },
+   "B cell": {
+    "n_significant": 4,
+    "n_tested": 4937,
+    "n_case_donors": 6,
+    "n_control_donors": 4,
+    "n_cells": 6177,
+    "top_list_size": 4
+   },
+   "Monocyte": {
+    "n_significant": 5,
+    "n_tested": 5151,
+    "n_case_donors": 6,
+    "n_control_donors": 4,
+    "n_cells": 18467,
+    "top_list_size": 5
+   },
+   "NK cell": {
+    "n_significant": 5,
+    "n_tested": 4928,
+    "n_case_donors": 6,
+    "n_control_donors": 4,
+    "n_cells": 9050,
+    "top_list_size": 5
+   },
+   "DC": {
+    "n_significant": 4,
+    "n_tested": 4621,
+    "n_case_donors": 6,
+    "n_control_donors": 4,
+    "n_cells": 1271,
+    "top_list_size": 4
+   }
+  },
+  "broadcast": "welch_bh_deg + top_deg_genes on the arm's condition bulk_counts (scoring genes, case|control donors), one list compared with every type's observed list (sle_broadcast.py broadcast_input_bulk); not part of the published benchmark"
+ },
+ "headline": {
+  "scsemi5rt": {
+   "comp_l1": 0.2706230707514841,
+   "comp_l1_obs_loo_null": 0.3545447454778543,
+   "precision": 0.9926666666666667,
+   "coverage": 0.5733333333333334,
+   "precision_dm": 0.518,
+   "coverage_dm": 0.22866666666666666,
+   "pb_r_median": 0.6097792936128839,
+   "pb_top1": 0.5,
+   "mono_pct_gen": 28.406666666666663,
+   "mono_pct_obs": 21.18723635141114,
+   "median_umi_gen": 465.0,
+   "median_umi_obs": 482.0,
+   "contrast_mean_cp10k_r_gen": 0.4215654020785214,
+   "contrast_pseudobulk_log1p_cpm_r_gen": 0.24736617626081037,
+   "contrast_mean_cp10k_r_broadcast": 0.5386553689123427,
+   "contrast_pseudobulk_log1p_cpm_r_broadcast": 0.6560793138377372,
+   "deg_top50_T": 0,
+   "deg_top50_B": 0,
+   "deg_top50_Mono": 0,
+   "deg_top50_NK": 0,
+   "deg_top50_DC": 0,
+   "deg_top50_sum": 0,
+   "deg_top50_max": 68,
+   "deg_top50_sum_4type": 0,
+   "deg_top50_sum_broadcast": 24,
+   "seconds": 5.3
+  }
+ },
+ "per_arm": {
+  "scsemi5rt": {
+   "arm": "scsemi5rt",
+   "npz": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/generated/giroux2022/fewshot_n5/draw1/scsemi5rt__raw.npz",
+   "npz_sha256": "d9d5463fd9a0d22606a1c22ff4c1eb6e932c0c2e311977d7a4bf91b350f99ebf",
+   "receipt_verified": true,
+   "generation": {
+    "checkpoint": null,
+    "checkpoint_sha256": null,
+    "input_condition_path": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/data/giroux2022/fewshot_n5/draw1/test_raw.npz",
+    "input_condition_sha256": "935e890f67fc789b9ecf47b5c90157bfe0a1983d6ffdb1ea2d198b6de9767f47",
+    "seed": 20260927,
+    "steps": null,
+    "support_readout": "upstream (values <= 10 normalized units zeroed), re-read from the saved per-target model; control of the read-out re-run (equals the official output up to z sampling noise)",
+    "weights": null
+   },
+   "condition_for_broadcast": {
+    "path": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/data/giroux2022/fewshot_n5/draw1/test_raw.npz",
+    "sha256": "935e890f67fc789b9ecf47b5c90157bfe0a1983d6ffdb1ea2d198b6de9767f47"
+   },
+   "poisson": {
+    "seed": 20260916,
+    "rule": "whole NPZ, rows in NPZ order",
+    "nnz_whole_npz": 1677542,
+    "median_library_whole_npz_panel": 465.0,
+    "n_rows_whole_npz": 15000
+   },
+   "annotation": {
+    "model_pt_sha256": "bf3655899e5abee435340fc008656c4a17342da0f30b70cfd43de8e9223e1930",
+    "zero_padded_model_genes": 0,
+    "label_counts": {
+     "T cell": 7581,
+     "Monocyte": 4261,
+     "NK cell": 1887,
+     "B cell": 827,
+     "Other": 363,
+     "DC": 81
+    },
+    "mean_l2_max_prob": 0.8384876937737068,
+    "predict_seconds": 3.6
+   },
+   "n_donors": 10,
+   "n_generated_cells": 15000,
+   "composition": {
+    "l1_mean": 0.2706230707514841,
+    "l1_median": 0.22261457141914498,
+    "l1_obs_loo_null_mean": 0.3545447454778543,
+    "l1_generated_loo_mean_vs_observed_mean": 0.42337062231285955,
+    "composition_top1": 0.5,
+    "mean_fraction_generated": {
+     "T cell": 0.5054,
+     "B cell": 0.055133,
+     "Monocyte": 0.284067,
+     "NK cell": 0.1258,
+     "DC": 0.0054,
+     "Other": 0.0242
+    },
+    "mean_fraction_observed": {
+     "T cell": 0.570331,
+     "B cell": 0.055565,
+     "Monocyte": 0.211872,
+     "NK cell": 0.09329,
+     "DC": 0.013132,
+     "Other": 0.05581
+    }
+   },
+   "pseudobulk": {
+    "n_genes": 5563,
+    "r_median": 0.6097792936128839,
+    "r_mean": 0.6138070200096901,
+    "top1": 0.5,
+    "n_top1": 5,
+    "own_rank_median": 2.0,
+    "chance_top1": 0.1
+   },
+   "precision_coverage": {
+    "precision": 0.9926666666666667,
+    "coverage": 0.5733333333333334,
+    "density": 1.0659333333333332,
+    "n_query": 1500,
+    "n_reference": 1500
+   },
+   "precision_coverage_depth_matched": {
+    "precision": 0.518,
+    "coverage": 0.22866666666666666,
+    "density": 0.18373333333333333,
+    "detected_median_before": {
+     "R1": 264.0,
+     "Q": 109.0
+    },
+    "detected_median_after": {
+     "R1": 109.0,
+     "Q": 109.0
+    }
+   },
+   "monocyte_pct": {
+    "generated_mean_over_donors": 28.406666666666663,
+    "observed_mean_over_donors": 21.18723635141114,
+    "generated_pooled_cells": 28.40666666666667
+   },
+   "median_umi_per_cell_scoring_genes": {
+    "generated_pooled": 465.0,
+    "observed_pooled": 482.0,
+    "generated_mean_of_donor_medians": 464.79998779296875,
+    "observed_mean_of_donor_medians": 499.35
+   },
+   "contrast": {
+    "mean_cp10k": {
+     "per_type": {
+      "T cell": {
+       "r_generated": 0.461007373754063,
+       "r_broadcast": 0.6850650089880258,
+       "partial_r_generated_given_broadcast": 0.20666962284446275,
+       "r_generated_vs_broadcast": 0.48016416220888797,
+       "n_genes": 2938
+      },
+      "B cell": {
+       "r_generated": 0.3267962363095575,
+       "r_broadcast": 0.43529553800990617,
+       "partial_r_generated_given_broadcast": 0.22592752282280693,
+       "r_generated_vs_broadcast": 0.30587248786425675,
+       "n_genes": 3121
+      },
+      "Monocyte": {
+       "r_generated": 0.4908860569310003,
+       "r_broadcast": 0.4929362363245346,
+       "partial_r_generated_given_broadcast": 0.3920381374930296,
+       "r_generated_vs_broadcast": 0.3468159100696941,
+       "n_genes": 3066
+      },
+      "NK cell": {
+       "r_generated": 0.4075719413194648,
+       "r_broadcast": 0.5413246923269038,
+       "partial_r_generated_given_broadcast": 0.22159569045728014,
+       "r_generated_vs_broadcast": 0.4446136040316459,
+       "n_genes": 3018
+      }
+     },
+     "mean_r_generated": 0.4215654020785214,
+     "mean_r_broadcast": 0.5386553689123427,
+     "mean_partial_r": 0.26155774340439486
+    },
+    "pseudobulk_log1p_cpm": {
+     "per_type": {
+      "T cell": {
+       "r_generated": 0.372771428325728,
+       "r_broadcast": 0.778305137206687,
+       "partial_r_generated_given_broadcast": 0.07725352233801083,
+       "r_generated_vs_broadcast": 0.42246438353816773,
+       "n_genes": 2938
+      },
+      "B cell": {
+       "r_generated": 0.14450465993784575,
+       "r_broadcast": 0.5433173120170087,
+       "partial_r_generated_given_broadcast": 0.01790483810602624,
+       "r_generated_vs_broadcast": 0.23910348470320145,
+       "n_genes": 3121
+      },
+      "Monocyte": {
+       "r_generated": 0.2155998404805981,
+       "r_broadcast": 0.6439809335079176,
+       "partial_r_generated_given_broadcast": 0.11921659066254596,
+       "r_generated_vs_broadcast": 0.19590885839874383,
+       "n_genes": 3066
+      },
+      "NK cell": {
+       "r_generated": 0.2565887762990696,
+       "r_broadcast": 0.6587138726193354,
+       "partial_r_generated_given_broadcast": 0.03486750803315499,
+       "r_generated_vs_broadcast": 0.35225643617782826,
+       "n_genes": 3018
+      }
+     },
+     "mean_r_generated": 0.24736617626081037,
+     "mean_r_broadcast": 0.6560793138377372,
+     "mean_partial_r": 0.06231061478493451
+    }
+   },
+   "deg_top50": {
+    "types_scored": [
+     "T cell",
+     "B cell",
+     "Monocyte",
+     "NK cell",
+     "DC"
+    ],
+    "types_not_scored": {},
+    "per_type": {
+     "T cell": {
+      "deg_scored": true,
+      "top50_overlap": 0,
+      "top50_jaccard": 0.0,
+      "obs_top_list_size": 50,
+      "gen_top_list_size": 0,
+      "top50_overlap_broadcast": 6,
+      "top50_jaccard_broadcast": 0.10526315789473684,
+      "n_cells_obs": 58362,
+      "n_cells_gen": 7581
+     },
+     "B cell": {
+      "deg_scored": true,
+      "top50_overlap": 0,
+      "top50_jaccard": 0.0,
+      "obs_top_list_size": 4,
+      "gen_top_list_size": 0,
+      "top50_overlap_broadcast": 4,
+      "top50_jaccard_broadcast": 0.3076923076923077,
+      "n_cells_obs": 6177,
+      "n_cells_gen": 827
+     },
+     "Monocyte": {
+      "deg_scored": true,
+      "top50_overlap": 0,
+      "top50_jaccard": 0.0,
+      "obs_top_list_size": 5,
+      "gen_top_list_size": 0,
+      "top50_overlap_broadcast": 5,
+      "top50_jaccard_broadcast": 0.38461538461538464,
+      "n_cells_obs": 18467,
+      "n_cells_gen": 4261
+     },
+     "NK cell": {
+      "deg_scored": true,
+      "top50_overlap": 0,
+      "top50_jaccard": 0.0,
+      "obs_top_list_size": 5,
+      "gen_top_list_size": 0,
+      "top50_overlap_broadcast": 5,
+      "top50_jaccard_broadcast": 0.38461538461538464,
+      "n_cells_obs": 9050,
+      "n_cells_gen": 1887
+     },
+     "DC": {
+      "deg_scored": true,
+      "top50_overlap": 0,
+      "top50_jaccard": 0.0,
+      "obs_top_list_size": 4,
+      "gen_top_list_size": 1,
+      "top50_overlap_broadcast": 4,
+      "top50_jaccard_broadcast": 0.3076923076923077,
+      "n_cells_obs": 1271,
+      "n_cells_gen": 81
+     }
+    },
+    "sum": 0,
+    "sum_4type": 0,
+    "max": 68,
+    "mean": 0.0,
+    "sum_broadcast": 24,
+    "sum_broadcast_4type": 20,
+    "broadcast_n_significant": 13,
+    "seconds": 0.02
+   },
+   "seconds": 5.3
+  }
+ },
+ "code_sha256": {
+  "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/tools/score_ft.py": "ec5f40b4cea6d5062d0138f3b934abb6465e34cbfeac869c8a0906924c033aec",
+  "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/tools/ft_common.py": "4a6366b0f22208cd58714fc0261b8840c643b9d4569704d560cfd93534cbc747",
+  "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/utility_studies_final6k_20260925/umap/cell_immicro_5arm/tools/stage_b_embed.py": "d056cfdf38fb0e5f478756f370174e993b75d0c973c48e02e21749175a4ca2da",
+  "/home/sj_server_1/PBISC/immun-pbmc-fm/bulk2sc-paper/scripts/evaluate_generated_vs_truth.py": "48b12d794782b840838853e93e0fc55fab00532c4b3698f4588e926b985d2881",
+  "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/utility_studies_final6k_20260925/immicro/tools/immicro_common.py": "c5c30e917148cc5e664e91ea30c11bbdff9b6d1f53a3e0a13e16b1917eff37e9",
+  "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/aifi_scanvi_rebuild_20260916/predict_scanvi.py": "482ee5dcf4d72b54bc36cbb6dcff62e5030bc7afc46cb5fb01eca73a82ee45a5",
+  "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/aifi_scanvi_rebuild_20260916/poisson_readout.py": "8bfa080942fbcdc89ae23b276524a1a83266fb8cbc56b9af279e50f82e6cde38"
+ },
+ "seconds": 5.8
+}