Archive 4 checkpoints (1-4)

.gitattributes

@@ -91,3 +91,4 @@ paper-releases/pbisc-diffusion/updates/20261005T175437Z-deck-supplementary-other
 paper-releases/pbisc-diffusion/updates/20261005T175437Z-deck-supplementary-other-tissues-review/diffusion/dataset/server_1/model8/presentations/2026-09-30--paper-summary--r00/PBISC-Diffusion_paper_summary_20260930.pptx filter=lfs diff=lfs merge=lfs -text
 paper-releases/pbisc-diffusion/updates/20261006T121425Z-fig1-architecture-plus-corpus/diffusion/dataset/server_1/manuscript/main.pdf filter=lfs diff=lfs merge=lfs -text
 paper-releases/pbisc-diffusion/updates/20261006T125143Z-supp-fig-s1-corpus/diffusion/dataset/server_1/manuscript/main.pdf filter=lfs diff=lfs merge=lfs -text
+paper-releases/pbisc-diffusion/updates/20261006T125143Z-supp-fig-s1-corpus/diffusion/dataset/server_1/manuscript/supplementary.pdf filter=lfs diff=lfs merge=lfs -text

paper-releases/pbisc-diffusion/updates/20261006T125143Z-supp-fig-s1-corpus/diffusion/dataset/server_1/manuscript/sections/supplementary_methods.tex

@@ -0,0 +1,51 @@
+\section{Supplementary Methods}
+\label{sec:supp-methods}
+
+\subsection{Pretraining corpus}
+The registered corpus holds 41,395,184 cells from 22 public datasets, 3,199 donors and 6,507 samples (Supplementary Fig.~\ref{fig:supp_corpus}). By platform, 74.8\% of cells come from 10x Genomics assays, 23.4\% from Parse, 1.7\% from ScaleBio and 0.1\% from Seq-Well. Most cells (88.5\%) come from donors without a recorded disease. This group includes 9.2 million cells from cytomegalovirus-seropositive donors, which the source metadata record as cytomegalovirus infection, and blood drawn after influenza vaccination. The rest come from autoimmune disease (7.1\%: systemic lupus erythematosus, rheumatoid arthritis, ACPA-positive individuals at risk of rheumatoid arthritis and juvenile dermatomyositis), infection (3.4\%: COVID-19, post-COVID-19 condition and influenza), hematologic disorders (1.0\%: myelodysplastic syndrome, myelodysplastic/myeloproliferative neoplasms and cytopenia) and drug reaction with eosinophilia and systemic symptoms ($<$0.1\%). One dataset is enriched for circulating hematopoietic progenitors and one for B cells. About a quarter of the cells (23.6\%) were cultured before sequencing, almost all of them in a 90-cytokine panel on peripheral blood mononuclear cells from 12 healthy donors. Where the source metadata lacked a condition or an assay, we took it from the GEO series records (GSE274680, GSE132802, GSE163121 and GSE314010) or from the data provider's description. Cell types come from each dataset's own labels, mapped to a shared two-level hierarchy: T cells 59.8\%, monocytes 17.8\%, B cells 9.0\%, NK cells 8.1\%, progenitors 2.9\% and dendritic cells 1.1\%, with 0.7\% unlabelled. These condition and cell-type labels describe the corpus only; the model never receives them.
+
+\begin{figure}[p]
+\centering
+% Authored at 180 mm like Fig. 1; 1.17\textwidth keeps the smallest labels near 4.5 pt and the legend on the page.
+\makebox[\textwidth][c]{\includegraphics[width=1.17\textwidth]{figures/figS_corpus.pdf}}
+\caption{\textbf{The pretraining corpus.}
+\textbf{a}, Shares of cells by split, by sample handling (in vitro: cultured or stimulated before sequencing) and by assay, and condition shares within each assay. Healthy includes cytomegalovirus-seropositive donors and post-vaccination draws.
+\textbf{b}, The 22 datasets: cells (log scale), donors, samples, and condition and cell-type (L1) shares.
+\textbf{c}, Cell types from each dataset's own labels, mapped to a shared two-level hierarchy: L1 shares and L2 cell counts (log scale).
+\textbf{d}, What each cell record carries. Condition and cell-type labels describe the corpus and are never model inputs.}
+\label{fig:supp_corpus}
+\end{figure}
+
+\subsection{Continuous noise schedule and reverse steps}
+Before clipping, the cumulative cosine coefficient was
+\begin{equation}
+ A(t)=\frac{\cos^2\!\left[\frac{t+0.008}{1.008}\frac{\pi}{2}\right]}{\cos^2\!\left[\frac{0.008}{1.008}\frac{\pi}{2}\right]}.
+\end{equation}
+The code clips $A(t)$ to $[10^{-12},1-10^{-12}]$. It then computes $\lambda(t)=\operatorname{clip}(\log A(t)-\log[1-A(t)],-20,20)$. The coefficients used by the model are $\alpha(t)=\sqrt{\operatorname{sigmoid}[\lambda(t)]}$ and $\sigma(t)=\sqrt{\operatorname{sigmoid}[-\lambda(t)]}$. The schedule uses float32 when model states use lower precision. Sinusoidal time features use 64 frequencies $\omega_k=10000^{-k/63}$. They join $\cos(2\pi t\omega_k)$ and $\sin(2\pi t\omega_k)$ for $k=0,\ldots,63$.
+
+For the denoiser prediction $\hat z_0=[\operatorname{clip}(\hat v,-1,1),\tanh(\ell/2)]$, reverse steps add no new noise. Between adjacent times $t>t'$, they use
+\begin{equation}
+ \hat\varepsilon=\frac{z_t-\alpha(t)\hat z_0}{\max[\sigma(t),10^{-8}]},\qquad
+ z_{t'}=\alpha(t')\hat z_0+\sigma(t')\hat\varepsilon.
+\end{equation}
+The grid has 51 endpoints $t_k=1-k/50$. It requires 50 denoiser calls. The final call, at $t=1/50$, returns its clean prediction directly. The previous clean prediction supplies self-conditioning during sampling. The final count-like output is
+\begin{equation}
+ \hat x_g=\left\{\exp\left[8\,\operatorname{clip}\left(\frac{\hat v_g+1}{2},0,1\right)\right]-1\right\}
+ \mathbb I(\ell_g>0)\,m_g.
+\end{equation}
+The code uses a condition check to set excluded values to exactly zero. It does not add Poisson or negative-binomial count draws. Nonzero generated values are therefore continuous, with a maximum of $\exp(8)-1$. A positive support prediction can still give a zero output when the clipped value reaches its lower endpoint. We therefore store separate masks for positive support and nonzero output.
+
+\subsection{Gene-module definitions}
+The IFN-core module contains \textit{ISG15, IFI6, IFI27, IFI35, IFI44, IFI44L, IFIT1, IFIT2, IFIT3, IFITM1, IFITM2, IFITM3, IRF7, IRF9, ISG20, MX1, MX2, OAS1, OAS2, OAS3, OASL, RSAD2, STAT1} and \textit{XAF1}. All 24 genes appear in the fixed gene list. The shared 1,945-gene SLE evaluation set lacks \textit{IRF9}, so its score uses 23 genes. The 11,589-gene set adds measured \textit{IRF9}.
+
+The B/plasmablast module in the fixed gene list contains \textit{MS4A1, CD19, CD27, CD38, CD79A, CD79B, BANK1, CD74, HLA-DRA, HLA-DRB1, TNFRSF13B, TNFRSF17, MZB1, XBP1, JCHAIN, PRDM1} and \textit{IRF4}. The original requested set also contained \textit{IGKC, IGHG1} and \textit{IGHG3}. These were absent from the fixed gene list, so the reported coverage uses 17 as its denominator. The shared gene set lacks \textit{CD27, TNFRSF17} and \textit{XBP1}. The 6,000- and 11,589-gene sets include measured \textit{CD27} and \textit{XBP1}, giving coverage of 16/17. Coverage means that genes can be tested. It does not mean that the model recovered their expression. These lists came from a saved pSS analysis for the related manuscript. The saved SLE records confirmed module names and used-gene counts but did not separately retain all original gene symbols.
+
+\subsection{How to read the single-recipe check}
+The $K=1000$ case was validation recipe \texttt{4ffc27115a79d477baa87e74}, sample \texttt{RV8959683}, donor \texttt{CV19-I-O-51}. Its original labels gave 614 monocytes, 227 T cells, 68 NK cells, 46 erythrocytes and 20 NKT cells. They also gave seven dendritic cells, seven platelets, six B/plasma cells, four hematopoietic stem/progenitor cells and one neutrophil. All 1,000 cells defined the target cell proportions. An earlier plot showed only 64 cells. We selected this case after an earlier inspection of nearest cells. It is not a random cohort sample. The reference and target sets shared no cells or donors. The independent classifier used 1,974 genes. This differed from both the 1,945-gene SLE set and the 1,750-gene set used in the earlier plot.
+
+We normalized reference profiles to 10,000 and applied $\log(1+x)$. We then averaged each class and applied $L_2$ normalization to its centroid. Query profiles used the same transform. Each label came from the class with the highest cosine similarity. An added check assigned Unknown when the top-two similarity difference was below 0.05. We separately reported cell proportions from original labels and from classifier-assigned real-cell labels. This showed errors in the reference labels assigned by the classifier. Neither comparison corrects unknown labeling errors on generated cells.
+
+\subsection{Saved records and limits of the tested models}
+The fixed manifest for the five checkpoints lists hashes for checkpoints and training settings. It also records the fixed gene identifiers, assay masks and per-cell seeds. Training checkpoints save both raw and EMA parameters. All reported reverse-sampling results use EMA parameters. The main analysis keeps Unknown cells and every other generated row. Each donor's generated file keeps outputs on the model's own gene set and exact mappings into the shared evaluation set. Source data include donor-selection rules, pseudobulks, per-gene tests, cell-proportion tables and embedding coordinates.
+
+The code uses separate transforms for gene selection, bulk conditioning, diffusion values and evaluation. Gene selection and evaluation use normalized-log expression. Conditioning uses bulk and depth features from the full gene list. Diffusion predicts clipped log raw counts. Replacing any of these transforms changes the experiment. The 20,097-gene result comes only from a 128-update pilot. The main results for the three gene sets cover only the stated training checkpoints. Neither the pilot nor the planned 16-slot schedule shows final performance after full-gene training has settled.

paper-releases/pbisc-diffusion/updates/20261006T125143Z-supp-fig-s1-corpus/diffusion/dataset/server_1/manuscript/source_data/corpus/README.md

@@ -0,0 +1,40 @@
+# Pretraining corpus tables (Figure 1 corpus panels)
+
+Built 2026-10-06 by `scripts/corpus/build_corpus_tables.py` (CPU only). `scripts/corpus/build_figure1_corpus.py`
+draws two figures from these files only: `figures/fig01_corpus.*`, the 38 mm strip that is Fig. 1c under the
+architecture (both authored at 180 mm; `sections/overview_figure.tex` places them at 1.05\textwidth), and
+`figures/figS_corpus.*`, the detailed version: Supplementary Fig. S1 with its text in Supplementary Methods S1.1
+(`sections/supplementary_methods.tex`, placed at 1.17\textwidth; overview, the 22 datasets, L1/L2 cell types, per-cell
+record). Methods (Training data) points to both.
+
+| File | Content |
+|---|---|
+| `corpus_summary.json` | totals and checks: 41,395,184 cells (41,270,779 base + 124,405 overlay), 22 datasets, 6,507 samples, split, condition, assay, cell-type totals, split-leakage check |
+| `corpus_datasets.csv` | one row per dataset: cells, samples, donors, assay, cells per split, cells per condition |
+| `corpus_samples.csv` | one row per registered sample (hashed `sample_key` only): platform, assay, split, majority condition |
+| `corpus_cellgroups.csv` | cells per sample x condition x handling (cell-level condition) |
+| `corpus_celltypes.csv` | cells per dataset x L1 x L2 (source labels mapped to the shared hierarchy) |
+| `corpus_assay_condition.csv` | cells per assay x condition |
+| `dataset_info.csv` | hand-written display names, accessions, enrichment and metadata notes |
+| `geo/` | GEO series matrices used to complete metadata missing from the source obs |
+
+Rules (all in the script): CMV-seropositive healthy donors (recorded as "cytomegalovirus infection") and
+post-vaccination draws count as Healthy. GSE274680 groups come from the GEO sample-title suffix (CON1/CON2
+control, ARI/LONG ACPA+ at risk, ERA early RA) and its donors from the GEO subject id. GSE132802 has one DRESS
+patient and one healthy volunteer. Parse 10M donors are healthy per the vendor description. In vitro = Parse
+cytokine panel (incl. PBS), GSE314010 (IL-36 or PBS, 7 h) and the GSE132802 cultured samples. Cell types are the
+block's harmonized source labels for base cells; overlay cells have no label ("Unlabelled").
+
+Split check: every sample sits in exactly one split (no sample-level overlap). At donor level,
+27 of 3,199 donors have samples in more than one split: 26 GSE274680 subjects, whose corpus donor key is the per-visit
+sample id (the GEO subject id is the person), and the GSE132802 patient, whose samples are separate time points and
+cultures. 38 val/test samples (659,112 cells; val 331,377, test 327,735) come from a person who also has
+training samples. Other datasets' donor ids were checked against the source person ids (e.g. Sound Life: 868 samples,
+96 people) and match.
+
+## Fig. 1c caption (as in `sections/overview_figure.tex`)
+
+**c**, Pretraining corpus: share of cells by assay (Other: Seq-Well and 10x of unreported chemistry), condition
+shares within each assay by disease family (Healthy includes CMV-seropositive donors, post-vaccination draws and
+cultured healthy-donor cells) and cell-type shares from each dataset's labels. The model never receives condition or
+cell-type labels.

paper-releases/pbisc-diffusion/updates/20261006T125143Z-supp-fig-s1-corpus/diffusion/dataset/server_1/manuscript/source_data/corpus/corpus_summary.json

@@ -0,0 +1,134 @@
+{
+ "cells": 41395184,
+ "base_cells": 41270779,
+ "overlay_cells": 124405,
+ "datasets": 22,
+ "samples": 6507,
+ "donors": 3199,
+ "split_cells": {
+  "test": 3892772,
+  "train": 33141380,
+  "val": 4361032
+ },
+ "split_samples": {
+  "train": 5232,
+  "val": 648,
+  "test": 627
+ },
+ "condition_cells": {
+  "ACPA+ at risk of RA": 1792129,
+  "COVID-19": 1279572,
+  "Cytopenia": 100147,
+  "DRESS": 25272,
+  "Healthy": 36639276,
+  "Influenza": 19233,
+  "JDM": 82666,
+  "MDS": 176233,
+  "MDS/MPN": 130268,
+  "Post-COVID-19": 97224,
+  "RA": 260504,
+  "SLE": 792660
+ },
+ "family_cells": {
+  "Autoimmune": 2927959,
+  "Drug reaction": 25272,
+  "Healthy": 36639276,
+  "Hematologic": 406648,
+  "Infection": 1396029
+ },
+ "handling_cells": {
+  "ex vivo": 31621421,
+  "in vitro": 9773763
+ },
+ "cmv_seropositive_healthy_cells": 9215690,
+ "assay_cells": {
+  "10x (version n.r.)": 64752,
+  "10x 3' (hashed)": 4117384,
+  "10x 3' v2": 2537693,
+  "10x 3' v3": 16999465,
+  "10x 5' v1": 1727948,
+  "10x 5' v2": 1727658,
+  "10x Flex": 3758514,
+  "10x v2 (3'/5')": 34051,
+  "Parse": 9697974,
+  "ScaleBio": 685024,
+  "Seq-Well": 44721
+ },
+ "platform_cells": {
+  "10x": 30967465,
+  "Parse": 9697974,
+  "ScaleBio": 685024,
+  "Seq-Well": 44721
+ },
+ "celltype_l1_cells": {
+  "B cell": 3744614,
+  "DC": 472529,
+  "Erythrocyte": 27504,
+  "Granulocyte": 1331,
+  "ILC": 19329,
+  "Mast cell": 65,
+  "Monocyte": 7373579,
+  "NK cell": 3369999,
+  "NKT cell": 2494,
+  "Neutrophil": 600,
+  "No label (overlay)": 124405,
+  "Platelet": 121626,
+  "Progenitor cell": 1206741,
+  "T cell": 24772208,
+  "__unknown__": 158160
+ },
+ "corpus_vs_block_donor_id_mismatch_rows": 0,
+ "samples_in_more_than_one_split": 0,
+ "donors_in_more_than_one_split": 27,
+ "cells_of_those_donors": 1743549,
+ "val_test_cells_whose_donor_has_train_samples": {
+  "test": 327735,
+  "val": 331377
+ },
+ "val_test_samples_whose_donor_has_train_samples": 38,
+ "donors_in_more_than_one_split_by_dataset": {
+  "061_A_": 26,
+  "300b_A": 1
+ },
+ "conditions_order": [
+  "Healthy",
+  "SLE",
+  "RA",
+  "ACPA+ at risk of RA",
+  "JDM",
+  "COVID-19",
+  "Post-COVID-19",
+  "Influenza",
+  "MDS",
+  "MDS/MPN",
+  "Cytopenia",
+  "DRESS"
+ ],
+ "condition_family": {
+  "Healthy": "Healthy",
+  "SLE": "Autoimmune",
+  "RA": "Autoimmune",
+  "ACPA+ at risk of RA": "Autoimmune",
+  "JDM": "Autoimmune",
+  "COVID-19": "Infection",
+  "Post-COVID-19": "Infection",
+  "Influenza": "Infection",
+  "MDS": "Hematologic",
+  "MDS/MPN": "Hematologic",
+  "Cytopenia": "Hematologic",
+  "DRESS": "Drug reaction"
+ },
+ "inputs": {
+  "corpus_obs": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_7_no_annotation_sh_hwang/data/e2_full_source/metadata/obs_sc.parquet",
+  "block_obs": "/home/sj_server_1/PBISC/PBISC-data/single-cell-data-block-blood/metadata/obs_sc.parquet",
+  "sample_index": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_7_no_annotation_sh_hwang/data/e2_full_index/metadata/sample_index.parquet",
+  "source_manifest": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_7_no_annotation_sh_hwang/data/e2_full_source/dataset_manifest.json",
+  "geo": [
+   "GSE132802_series_matrix.txt.gz",
+   "GSE163121_series_matrix.txt.gz",
+   "GSE274680_series_matrix.txt.gz",
+   "GSE314010_series_matrix.txt.gz"
+  ]
+ },
+ "script": "/home/sj_server_1/PBISC/paper-release-20260928/manuscript/scripts/corpus/build_corpus_tables.py"
+}

paper-releases/pbisc-diffusion/updates/20261006T125143Z-supp-fig-s1-corpus/diffusion/dataset/server_1/manuscript/supplementary.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:edf373e44050f63623b48a73f07cca90dbe15ab5808e1b03eabebbd001543d00
+size 2516919