Datasets:

junidude14
/

PBISC-data

paper-releases/pbisc-diffusion/updates/20261002T124820Z-crosscohort-liu-default/diffusion/dataset/server_1/model8/plans/paired_crosscohort_20261002/scores/giroux2022/loco_lora16_K1000/README.md ADDED Viewed

	@@ -0,0 +1,73 @@

+# score_ft.py output: giroux2022, 24 scored donors
+Arms: loco_lora16_K1000.  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 24 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) (3600 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 0.9981, 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
+   0.9981, 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 = 11) vs control groups ['mild'] (n = 13); 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 11 + 13
+   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).

	@@ -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
+zs_K1000,T cell,mean_cp10k,3021,0.19571663549808774,0.22039608693345034,0.1482708208152166,0.2532010443016586,18,15,18,15,59396,14510,,,,,,,,,,,
+zs_K1000,T cell,pseudobulk_log1p_cpm,3021,0.10954524145016434,0.27145331834905573,0.0776578066867918,0.13056831473567765,18,15,18,15,59396,14510,,,,,,,,,,,
+zs_K1000,B cell,mean_cp10k,3245,-0.08908110897621051,0.13326750229861392,-0.1077651779992899,0.1265441007184561,18,15,18,15,7162,2617,,,,,,,,,,,
+zs_K1000,B cell,pseudobulk_log1p_cpm,3245,-0.09982618477043084,0.01993963216355321,-0.10131847867152856,0.06355863652163926,18,15,18,15,7162,2617,,,,,,,,,,,
+zs_K1000,Monocyte,mean_cp10k,3154,0.2036215726172367,0.3983317004681723,0.09809050580716451,0.295391531877229,18,15,18,15,34457,26202,,,,,,,,,,,
+zs_K1000,Monocyte,pseudobulk_log1p_cpm,3154,0.13168819603510043,0.35676796217145185,0.08453324573079685,0.15027874254241963,18,15,18,15,34457,26202,,,,,,,,,,,
+zs_K1000,NK cell,mean_cp10k,3051,-0.010458282526034986,0.20750320008350984,-0.04661204907651572,0.1662837651651097,18,15,18,15,18147,1074,,,,,,,,,,,
+zs_K1000,NK cell,pseudobulk_log1p_cpm,3051,0.034624328122922286,0.10364196133517908,0.022351771627865346,0.12115452118158426,18,15,18,15,18147,1074,,,,,,,,,,,
+zs_K1000,T cell,welch_top50_deg,5752,,,,,18,15,18,15,59396,14510,True,0,0.0,50,1,1897,1,0,0.0,15,15
+zs_K1000,B cell,welch_top50_deg,5752,,,,,18,15,18,15,7162,2617,True,0,0.0,50,2,73,2,1,0.015625,15,15
+zs_K1000,Monocyte,welch_top50_deg,5752,,,,,18,15,18,15,34457,26202,True,0,0.0,50,5,346,5,2,0.031746031746031744,15,15
+zs_K1000,NK cell,welch_top50_deg,5752,,,,,18,15,18,15,18147,1074,True,0,0.0,50,1,396,1,1,0.015625,15,15
+zs_K1000,DC,welch_top50_deg,5752,,,,,16,15,18,15,1755,240,True,0,0.0,50,0,138,0,0,0.0,15,15

paper-releases/pbisc-diffusion/updates/20261002T124820Z-crosscohort-liu-default/diffusion/dataset/server_1/model8/plans/paired_crosscohort_20261002/scores/ye2026_sepsis/loco_lora16_K1000/contrast.csv ADDED Viewed

	@@ -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
+loco_lora16_K1000,T cell,mean_cp10k,3258,-0.011975171459846714,0.2548893770824626,-0.05122236503031578,0.1452781368297155,61,21,61,21,298117,43422,,,,,,,,,,,
+loco_lora16_K1000,T cell,pseudobulk_log1p_cpm,3258,0.04800011269924342,0.2164745523777111,0.0015655672204354047,0.21483983940437876,61,21,61,21,298117,43422,,,,,,,,,,,
+loco_lora16_K1000,B cell,mean_cp10k,3310,0.16157610893841604,0.27458639801653467,0.10525937099016786,0.22968585775784886,61,21,61,21,99729,29702,,,,,,,,,,,
+loco_lora16_K1000,B cell,pseudobulk_log1p_cpm,3310,0.2821712145687276,0.2711145517540278,0.22054611458977494,0.2918619927416837,61,21,61,21,99729,29702,,,,,,,,,,,
+loco_lora16_K1000,Monocyte,mean_cp10k,3289,0.21645254552397006,0.5481103923782986,0.10304899373389625,0.2423436168446589,61,21,61,21,111269,24672,,,,,,,,,,,
+loco_lora16_K1000,Monocyte,pseudobulk_log1p_cpm,3289,0.1512310645029765,0.6203629305570988,-0.03342966230326073,0.2842989364079799,61,21,61,21,111269,24672,,,,,,,,,,,
+loco_lora16_K1000,NK cell,mean_cp10k,3240,0.0004305268013675477,0.1861970975708186,-0.02278747175940871,0.12166239617831853,61,21,61,21,47832,18981,,,,,,,,,,,
+loco_lora16_K1000,NK cell,pseudobulk_log1p_cpm,3240,0.029464650213740368,0.05592238177293225,0.0221225534008507,0.13555562487052558,61,21,61,21,47832,18981,,,,,,,,,,,
+loco_lora16_K1000,T cell,welch_top50_deg,5975,,,,,61,21,61,21,298117,43422,True,1,0.019230769230769232,50,3,1578,3,3,0.030927835051546393,50,1378
+loco_lora16_K1000,B cell,welch_top50_deg,5975,,,,,61,21,61,21,99729,29702,True,1,0.010101010101010102,50,50,1102,429,2,0.02040816326530612,50,1378
+loco_lora16_K1000,Monocyte,welch_top50_deg,5975,,,,,61,21,61,21,111269,24672,True,5,0.05263157894736842,50,50,2459,177,13,0.14942528735632185,50,1378
+loco_lora16_K1000,NK cell,welch_top50_deg,5975,,,,,61,21,61,21,47832,18981,True,0,0.0,50,1,1235,1,2,0.02040816326530612,50,1378
+loco_lora16_K1000,DC,welch_top50_deg,5975,,,,,57,21,61,21,3300,3737,True,0,0.0,50,3,1572,3,1,0.010101010101010102,50,1378

	@@ -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
+zs_K1000,T cell,mean_cp10k,3258,-0.10432230143312773,0.2548893770824626,-0.17298372266935502,0.2294525413871452,61,21,61,21,298117,43713,,,,,,,,,,,
+zs_K1000,T cell,pseudobulk_log1p_cpm,3258,-0.026301310225745313,0.2164745523777111,-0.0886971581714242,0.2642967617857989,61,21,61,21,298117,43713,,,,,,,,,,,
+zs_K1000,B cell,mean_cp10k,3310,0.047437779482892545,0.27458639801653467,-0.015553323632773097,0.2258194826100485,61,21,61,21,99729,18529,,,,,,,,,,,
+zs_K1000,B cell,pseudobulk_log1p_cpm,3310,0.09858083558441715,0.2711145517540278,0.03901229562465431,0.22878001258534544,61,21,61,21,99729,18529,,,,,,,,,,,
+zs_K1000,Monocyte,mean_cp10k,3289,0.049849833404431006,0.5481103923782986,-0.0829739879467094,0.21461492174973537,61,21,61,21,111269,45231,,,,,,,,,,,
+zs_K1000,Monocyte,pseudobulk_log1p_cpm,3289,0.18929013682344356,0.6203629305570988,-0.05740162287711388,0.3724778103826463,61,21,61,21,111269,45231,,,,,,,,,,,
+zs_K1000,NK cell,mean_cp10k,3240,-0.08036052644753712,0.1861970975708186,-0.14059337337126476,0.2805016269078411,61,21,61,21,47832,11319,,,,,,,,,,,
+zs_K1000,NK cell,pseudobulk_log1p_cpm,3240,-0.06394830647645451,0.05592238177293225,-0.08252801674133732,0.27367752588949024,61,21,61,21,47832,11319,,,,,,,,,,,
+zs_K1000,T cell,welch_top50_deg,5975,,,,,61,21,61,21,298117,43713,True,3,0.030927835051546393,50,50,1578,749,3,0.030927835051546393,50,1378
+zs_K1000,B cell,welch_top50_deg,5975,,,,,61,21,61,21,99729,18529,True,0,0.0,50,50,1102,548,2,0.02040816326530612,50,1378
+zs_K1000,Monocyte,welch_top50_deg,5975,,,,,61,21,61,21,111269,45231,True,4,0.041666666666666664,50,50,2459,946,13,0.14942528735632185,50,1378
+zs_K1000,NK cell,welch_top50_deg,5975,,,,,61,21,61,21,47832,11319,True,1,0.010101010101010102,50,50,1235,302,2,0.02040816326530612,50,1378
+zs_K1000,DC,welch_top50_deg,5975,,,,,57,21,60,21,3300,1159,True,1,0.010416666666666666,50,47,1572,47,1,0.010101010101010102,50,1378