Archive 4 checkpoints (1-4)
Browse files- paper-releases/pbisc-diffusion/updates/20261008T022410Z-deconv-scsemi-deck/diffusion/dataset/server_1/model8/plans/bayesprism_paired_20261006/logs/cx_hires_ye2026_sepsis_fewshot3.log +15 -0
- paper-releases/pbisc-diffusion/updates/20261008T022410Z-deconv-scsemi-deck/diffusion/dataset/server_1/model8/plans/bayesprism_paired_20261006/logs/cx_hires_ye2026_sepsis_fewshot4.log +15 -0
- paper-releases/pbisc-diffusion/updates/20261008T022410Z-deconv-scsemi-deck/diffusion/dataset/server_1/model8/plans/bayesprism_paired_20261006/logs/finalize_20261006T155858.log +103 -0
- paper-releases/pbisc-diffusion/updates/20261008T022410Z-deconv-scsemi-deck/diffusion/dataset/server_1/model8/plans/bayesprism_paired_20261006/logs/finalize_cx_20261008T105628.log +126 -0
paper-releases/pbisc-diffusion/updates/20261008T022410Z-deconv-scsemi-deck/diffusion/dataset/server_1/model8/plans/bayesprism_paired_20261006/logs/cx_hires_ye2026_sepsis_fewshot3.log
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
>Running CIBERSORTx high-resolution GEP imputation...
|
| 2 |
+
>[Options] username: [REDACTED]
|
| 3 |
+
>[Options] token: [REDACTED]
|
| 4 |
+
>[Options] mixture: mixture.txt
|
| 5 |
+
>[Options] sigmatrix: sigmatrix.txt
|
| 6 |
+
>[Options] cibresults: fractions_results.txt
|
| 7 |
+
>[Options] rmbatchSmode: TRUE
|
| 8 |
+
>[Options] sourceGEPs: source_geps.txt
|
| 9 |
+
>[Options] subsetgenes: hires_genes.txt
|
| 10 |
+
>[Options] threads: 4
|
| 11 |
+
>Loaded 77 mixture samples, 5975 genes, and 6 cell subsets...
|
| 12 |
+
>Window size adaptively set to 24
|
| 13 |
+
>Imputing high-resolution cell type GEPs...done.
|
| 14 |
+
>Writing output to disk ...done.
|
| 15 |
+
>Running time (sec): 1108
|
paper-releases/pbisc-diffusion/updates/20261008T022410Z-deconv-scsemi-deck/diffusion/dataset/server_1/model8/plans/bayesprism_paired_20261006/logs/cx_hires_ye2026_sepsis_fewshot4.log
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
>Running CIBERSORTx high-resolution GEP imputation...
|
| 2 |
+
>[Options] username: [REDACTED]
|
| 3 |
+
>[Options] token: [REDACTED]
|
| 4 |
+
>[Options] mixture: mixture.txt
|
| 5 |
+
>[Options] sigmatrix: sigmatrix.txt
|
| 6 |
+
>[Options] cibresults: fractions_results.txt
|
| 7 |
+
>[Options] rmbatchSmode: TRUE
|
| 8 |
+
>[Options] sourceGEPs: source_geps.txt
|
| 9 |
+
>[Options] subsetgenes: hires_genes.txt
|
| 10 |
+
>[Options] threads: 4
|
| 11 |
+
>Loaded 77 mixture samples, 5975 genes, and 6 cell subsets...
|
| 12 |
+
>Window size adaptively set to 24
|
| 13 |
+
>Imputing high-resolution cell type GEPs...done.
|
| 14 |
+
>Writing output to disk ...done.
|
| 15 |
+
>Running time (sec): 1077
|
paper-releases/pbisc-diffusion/updates/20261008T022410Z-deconv-scsemi-deck/diffusion/dataset/server_1/model8/plans/bayesprism_paired_20261006/logs/finalize_20261006T155858.log
ADDED
|
@@ -0,0 +1,103 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2026-10-06T18:49:00 lanes finished
|
| 2 |
+
Traceback (most recent call last):
|
| 3 |
+
File "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/bayesprism_paired_20261006/tools/score_bp.py", line 268, in <module>
|
| 4 |
+
sys.exit(main())
|
| 5 |
+
^^^^^^
|
| 6 |
+
File "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/bayesprism_paired_20261006/tools/score_bp.py", line 178, in main
|
| 7 |
+
contrast = {"enabled": True, "case": np.asarray(cd["case"], bool), "ctrl": np.asarray(cd["ctrl"], bool)}
|
| 8 |
+
~~^^^^^^^^
|
| 9 |
+
KeyError: 'case'
|
| 10 |
+
2026-10-06T18:49:01 WARN: immicro fewshot 0
|
| 11 |
+
Traceback (most recent call last):
|
| 12 |
+
File "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/bayesprism_paired_20261006/tools/score_bp.py", line 268, in <module>
|
| 13 |
+
sys.exit(main())
|
| 14 |
+
^^^^^^
|
| 15 |
+
File "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/bayesprism_paired_20261006/tools/score_bp.py", line 178, in main
|
| 16 |
+
contrast = {"enabled": True, "case": np.asarray(cd["case"], bool), "ctrl": np.asarray(cd["ctrl"], bool)}
|
| 17 |
+
~~^^^^^^^^
|
| 18 |
+
KeyError: 'case'
|
| 19 |
+
2026-10-06T18:49:02 WARN: immicro fewshot 1
|
| 20 |
+
Traceback (most recent call last):
|
| 21 |
+
File "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/bayesprism_paired_20261006/tools/score_bp.py", line 268, in <module>
|
| 22 |
+
sys.exit(main())
|
| 23 |
+
^^^^^^
|
| 24 |
+
File "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/bayesprism_paired_20261006/tools/score_bp.py", line 178, in main
|
| 25 |
+
contrast = {"enabled": True, "case": np.asarray(cd["case"], bool), "ctrl": np.asarray(cd["ctrl"], bool)}
|
| 26 |
+
~~^^^^^^^^
|
| 27 |
+
KeyError: 'case'
|
| 28 |
+
2026-10-06T18:49:03 WARN: immicro fewshot 2
|
| 29 |
+
Traceback (most recent call last):
|
| 30 |
+
File "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/bayesprism_paired_20261006/tools/score_bp.py", line 268, in <module>
|
| 31 |
+
sys.exit(main())
|
| 32 |
+
^^^^^^
|
| 33 |
+
File "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/bayesprism_paired_20261006/tools/score_bp.py", line 178, in main
|
| 34 |
+
contrast = {"enabled": True, "case": np.asarray(cd["case"], bool), "ctrl": np.asarray(cd["ctrl"], bool)}
|
| 35 |
+
~~^^^^^^^^
|
| 36 |
+
KeyError: 'case'
|
| 37 |
+
2026-10-06T18:49:04 WARN: immicro fewshot 3
|
| 38 |
+
Traceback (most recent call last):
|
| 39 |
+
File "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/bayesprism_paired_20261006/tools/score_bp.py", line 268, in <module>
|
| 40 |
+
sys.exit(main())
|
| 41 |
+
^^^^^^
|
| 42 |
+
File "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/bayesprism_paired_20261006/tools/score_bp.py", line 178, in main
|
| 43 |
+
contrast = {"enabled": True, "case": np.asarray(cd["case"], bool), "ctrl": np.asarray(cd["ctrl"], bool)}
|
| 44 |
+
~~^^^^^^^^
|
| 45 |
+
KeyError: 'case'
|
| 46 |
+
2026-10-06T18:49:05 WARN: immicro fewshot 4
|
| 47 |
+
Traceback (most recent call last):
|
| 48 |
+
File "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/bayesprism_paired_20261006/tools/score_bp.py", line 268, in <module>
|
| 49 |
+
sys.exit(main())
|
| 50 |
+
^^^^^^
|
| 51 |
+
File "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/bayesprism_paired_20261006/tools/score_bp.py", line 178, in main
|
| 52 |
+
contrast = {"enabled": True, "case": np.asarray(cd["case"], bool), "ctrl": np.asarray(cd["ctrl"], bool)}
|
| 53 |
+
~~^^^^^^^^
|
| 54 |
+
KeyError: 'case'
|
| 55 |
+
2026-10-06T18:49:06 WARN: immicro cv
|
| 56 |
+
{"dataset": "ye2026_sepsis", "comparison": "fewshot", "draw": 0, "bayesprism": {"comp_l1": 1.1392577455707993, "comp_l1_by_variant": {"final_cells": 1.1392577455707993, "first_cells": 0.6220990907002514, "final_mrna": 1.1152808734771347, "first_mrna": 0.6222838313901287}, "logfc_r_mean_cp10k": 0.2773581324391346, "partial_r_mean_cp10k": 0.140270759674815, "logfc_r_mean_cp10k_covered_genes": 0.2895072138637591, "deg_top50_sum": 34, "deg_top50_sum_covered_genes": 34, "per_type_logfc_r": {"T cell": 0.1762, "B cell": 0.2161, "Monocyte": 0.599, "NK cell": 0.1181}}, "lora": {"comp_l1": 0.4723364713369038, "logfc_r_mean_cp10k": 0.4359905652333528, "deg_top50_sum": 7, "partial_r_mean_cp10k": 0.4060106147529024, "coverage": 0.7058874458874459}, "broadcast": {"logfc_r_mean_cp10k": 0.3359901153517466, "logfc_r_mean_cp10k_covered_genes": 0.27827433682632724, "deg_top50_sum": 23}}
|
| 57 |
+
{"dataset": "ye2026_sepsis", "comparison": "fewshot", "draw": 1, "bayesprism": {"comp_l1": 1.181910168171711, "comp_l1_by_variant": {"final_cells": 1.181910168171711, "first_cells": 0.9197332138543787, "final_mrna": 1.1809189691067659, "first_mrna": 0.9290749025987671}, "logfc_r_mean_cp10k": 0.18585702234665386, "partial_r_mean_cp10k": 0.07555845000011678, "logfc_r_mean_cp10k_covered_genes": 0.19277961723852588, "deg_top50_sum": 32, "deg_top50_sum_covered_genes": 32, "per_type_logfc_r": {"T cell": 0.0623, "B cell": 0.0802, "Monocyte": 0.5748, "NK cell": 0.026}}, "lora": {"comp_l1": 0.4725655492172997, "logfc_r_mean_cp10k": 0.08123703391754132, "deg_top50_sum": 16, "partial_r_mean_cp10k": -0.018935679613692524, "coverage": 0.6619047619047619}, "broadcast": {"logfc_r_mean_cp10k": 0.27941436238089834, "logfc_r_mean_cp10k_covered_genes": 0.220702686320993, "deg_top50_sum": 20}}
|
| 58 |
+
{"dataset": "ye2026_sepsis", "comparison": "fewshot", "draw": 2, "bayesprism": {"comp_l1": 1.1756539131021078, "comp_l1_by_variant": {"final_cells": 1.1756539131021078, "first_cells": 0.8327452885627836, "final_mrna": 1.1413766519413018, "first_mrna": 0.8258016263679072}, "logfc_r_mean_cp10k": 0.28745787646823584, "partial_r_mean_cp10k": 0.15293173894860773, "logfc_r_mean_cp10k_covered_genes": 0.29933387653200805, "deg_top50_sum": 25, "deg_top50_sum_covered_genes": 25, "per_type_logfc_r": {"T cell": 0.1927, "B cell": 0.209, "Monocyte": 0.5632, "NK cell": 0.1849}}, "lora": {"comp_l1": 0.4986968896699433, "logfc_r_mean_cp10k": 0.32721881191208096, "deg_top50_sum": 26, "partial_r_mean_cp10k": 0.31447561738686153, "coverage": 0.7606060606060606}, "broadcast": {"logfc_r_mean_cp10k": 0.3386821732452143, "logfc_r_mean_cp10k_covered_genes": 0.2912856085705393, "deg_top50_sum": 18}}
|
| 59 |
+
{"dataset": "ye2026_sepsis", "comparison": "fewshot", "draw": 3, "bayesprism": {"comp_l1": 1.2376129822519755, "comp_l1_by_variant": {"final_cells": 1.2376129822519755, "first_cells": 1.0438777591557735, "final_mrna": 1.1442940011539209, "first_mrna": 0.9665424686618387}, "logfc_r_mean_cp10k": 0.2885661552108418, "partial_r_mean_cp10k": 0.1626244185339312, "logfc_r_mean_cp10k_covered_genes": 0.3011671152115622, "deg_top50_sum": 20, "deg_top50_sum_covered_genes": 20, "per_type_logfc_r": {"T cell": 0.1693, "B cell": 0.2378, "Monocyte": 0.566, "NK cell": 0.1811}}, "lora": {"comp_l1": 0.4968788955478774, "logfc_r_mean_cp10k": 0.4108841123350609, "deg_top50_sum": 17, "partial_r_mean_cp10k": 0.3622128751991822, "coverage": 0.6553246753246753}, "broadcast": {"logfc_r_mean_cp10k": 0.3264466405549955, "logfc_r_mean_cp10k_covered_genes": 0.27365588419521647, "deg_top50_sum": 19}}
|
| 60 |
+
{"dataset": "ye2026_sepsis", "comparison": "fewshot", "draw": 4, "bayesprism": {"comp_l1": 1.1503124470269723, "comp_l1_by_variant": {"final_cells": 1.1503124470269723, "first_cells": 0.7707407514452587, "final_mrna": 1.117261442619167, "first_mrna": 0.7573736778194846}, "logfc_r_mean_cp10k": 0.2568730863808132, "partial_r_mean_cp10k": 0.11594592401612624, "logfc_r_mean_cp10k_covered_genes": 0.2684078662410606, "deg_top50_sum": 34, "deg_top50_sum_covered_genes": 34, "per_type_logfc_r": {"T cell": 0.1236, "B cell": 0.1979, "Monocyte": 0.5155, "NK cell": 0.1905}}, "lora": {"comp_l1": 0.39049012270725986, "logfc_r_mean_cp10k": 0.49526136963540107, "deg_top50_sum": 16, "partial_r_mean_cp10k": 0.4597246910197378, "coverage": 0.687012987012987}, "broadcast": {"logfc_r_mean_cp10k": 0.3195942137906433, "logfc_r_mean_cp10k_covered_genes": 0.26232772053396347, "deg_top50_sum": 20}}
|
| 61 |
+
condition NPZ not found: /home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/generated/ye2026_sepsis/pooled/conditions_pooled_raw.npz (pass --condition ARM=PATH or --no-contrast)
|
| 62 |
+
2026-10-06T18:49:17 WARN: ye2026_sepsis cv
|
| 63 |
+
{"dataset": "ji2025_chili", "comparison": "fewshot", "draw": 0, "bayesprism": {"comp_l1": 1.0881064322609375, "comp_l1_by_variant": {"final_cells": 1.0881064322609375, "first_cells": 1.035522254704717, "final_mrna": 1.1230510137666498, "first_mrna": 1.0617398622381342}, "logfc_r_mean_cp10k": 0.2100217096465254, "partial_r_mean_cp10k": 0.05705121990314871, "logfc_r_mean_cp10k_covered_genes": 0.21476502864578612, "deg_top50_sum": 0, "deg_top50_sum_covered_genes": 0, "per_type_logfc_r": {"T cell": 0.2398, "B cell": 0.1065, "Monocyte": 0.37, "NK cell": 0.1238}}, "lora": {"comp_l1": 0.46942970758493824, "logfc_r_mean_cp10k": 0.38697985726899736, "deg_top50_sum": 0, "partial_r_mean_cp10k": 0.36588903106531123, "coverage": 0.6992}, "broadcast": {"logfc_r_mean_cp10k": 0.21311861411338662, "logfc_r_mean_cp10k_covered_genes": 0.20115944437037211, "deg_top50_sum": 0}}
|
| 64 |
+
{"dataset": "ji2025_chili", "comparison": "fewshot", "draw": 1, "bayesprism": {"comp_l1": 1.0852947355832807, "comp_l1_by_variant": {"final_cells": 1.0852947355832807, "first_cells": 1.0227093442188013, "final_mrna": 1.1177013725310994, "first_mrna": 1.0505150333470727}, "logfc_r_mean_cp10k": 0.18355806038771155, "partial_r_mean_cp10k": 0.04052953941950543, "logfc_r_mean_cp10k_covered_genes": 0.18643910243579465, "deg_top50_sum": 2, "deg_top50_sum_covered_genes": 2, "per_type_logfc_r": {"T cell": 0.1439, "B cell": 0.0126, "Monocyte": 0.4472, "NK cell": 0.1305}}, "lora": {"comp_l1": 0.3891842501829222, "logfc_r_mean_cp10k": 0.27568480061697936, "deg_top50_sum": 1, "partial_r_mean_cp10k": 0.2420147625491504, "coverage": 0.8551282051282051}, "broadcast": {"logfc_r_mean_cp10k": 0.20791984755382747, "logfc_r_mean_cp10k_covered_genes": 0.20072022902540193, "deg_top50_sum": 0}}
|
| 65 |
+
{"dataset": "ji2025_chili", "comparison": "fewshot", "draw": 2, "bayesprism": {"comp_l1": 1.1194003390045488, "comp_l1_by_variant": {"final_cells": 1.1194003390045488, "first_cells": 1.0938459550012145, "final_mrna": 1.1750518444226243, "first_mrna": 1.1382535431910876}, "logfc_r_mean_cp10k": 0.20290991366474348, "partial_r_mean_cp10k": 0.009427337713952683, "logfc_r_mean_cp10k_covered_genes": 0.20487837856844668, "deg_top50_sum": 2, "deg_top50_sum_covered_genes": 2, "per_type_logfc_r": {"T cell": 0.1557, "B cell": 0.1013, "Monocyte": 0.3905, "NK cell": 0.1642}}, "lora": {"comp_l1": 0.5825794078861584, "logfc_r_mean_cp10k": 0.28938101620694295, "deg_top50_sum": 0, "partial_r_mean_cp10k": 0.24803917247992743, "coverage": 0.5297435897435897}, "broadcast": {"logfc_r_mean_cp10k": 0.263723888008903, "logfc_r_mean_cp10k_covered_genes": 0.24932988554722285, "deg_top50_sum": 5}}
|
| 66 |
+
{"dataset": "ji2025_chili", "comparison": "fewshot", "draw": 3, "bayesprism": {"comp_l1": 1.0780235525884094, "comp_l1_by_variant": {"final_cells": 1.0780235525884094, "first_cells": 1.0285215433682688, "final_mrna": 1.1394889517460378, "first_mrna": 1.0797575172345943}, "logfc_r_mean_cp10k": 0.22068415612787792, "partial_r_mean_cp10k": 0.044297362609563254, "logfc_r_mean_cp10k_covered_genes": 0.2223200818882785, "deg_top50_sum": 0, "deg_top50_sum_covered_genes": 0, "per_type_logfc_r": {"T cell": 0.1208, "B cell": 0.1784, "Monocyte": 0.4127, "NK cell": 0.1708}}, "lora": {"comp_l1": 0.4487906027467767, "logfc_r_mean_cp10k": 0.4410983924408845, "deg_top50_sum": 0, "partial_r_mean_cp10k": 0.39982215110012487, "coverage": 0.7586666666666667}, "broadcast": {"logfc_r_mean_cp10k": 0.2577780039471977, "logfc_r_mean_cp10k_covered_genes": 0.24577219938281142, "deg_top50_sum": 0}}
|
| 67 |
+
{"dataset": "ji2025_chili", "comparison": "fewshot", "draw": 4, "bayesprism": {"comp_l1": 1.0520912287908002, "comp_l1_by_variant": {"final_cells": 1.0520912287908002, "first_cells": 1.019637997458479, "final_mrna": 1.111736305498087, "first_mrna": 1.0882082584361317}, "logfc_r_mean_cp10k": 0.21636260390760662, "partial_r_mean_cp10k": 0.05907257486255502, "logfc_r_mean_cp10k_covered_genes": 0.22009966491136165, "deg_top50_sum": 4, "deg_top50_sum_covered_genes": 4, "per_type_logfc_r": {"T cell": 0.2503, "B cell": 0.1007, "Monocyte": 0.3806, "NK cell": 0.1337}}, "lora": {"comp_l1": 0.6754540168870949, "logfc_r_mean_cp10k": 0.2725492074488929, "deg_top50_sum": 0, "partial_r_mean_cp10k": 0.2467295164617603, "coverage": 0.6936111111111111}, "broadcast": {"logfc_r_mean_cp10k": 0.2254541110690231, "logfc_r_mean_cp10k_covered_genes": 0.20941395571219967, "deg_top50_sum": 0}}
|
| 68 |
+
{"dataset": "ji2025_chili", "comparison": "cv", "draw": null, "bayesprism": {"comp_l1": 1.0978294301798797, "comp_l1_by_variant": {"final_cells": 1.0978294301798797, "first_cells": 1.0647439954894382, "final_mrna": 1.1342158386624934, "first_mrna": 1.0998312499793417}, "logfc_r_mean_cp10k": 0.24055359091687248, "partial_r_mean_cp10k": 0.07660192688042304, "logfc_r_mean_cp10k_covered_genes": 0.2505351616532235, "deg_top50_sum": 4, "deg_top50_sum_covered_genes": 4, "per_type_logfc_r": {"T cell": 0.1821, "B cell": 0.2198, "Monocyte": 0.4224, "NK cell": 0.1379}}, "lora": {"comp_l1": 0.5044992874562503, "logfc_r_mean_cp10k": 0.47928798340225065, "deg_top50_sum": 15, "partial_r_mean_cp10k": 0.45204372964017614, "coverage": 0.8569696969696969}, "broadcast": {"logfc_r_mean_cp10k": 0.2398746224459366, "logfc_r_mean_cp10k_covered_genes": 0.22569844870512668, "deg_top50_sum": 4}}
|
| 69 |
+
{"dataset": "imperial_eskd", "comparison": "fewshot", "draw": 0, "bayesprism": {"comp_l1": 1.2775618819907144, "comp_l1_by_variant": {"final_cells": 1.2775618819907144, "first_cells": 1.206422325810398, "final_mrna": 1.2858338337171236, "first_mrna": 1.1879496593758736}, "logfc_r_mean_cp10k": 0.6599409341291975, "partial_r_mean_cp10k": 0.32709876951819394, "logfc_r_mean_cp10k_covered_genes": 0.6647928060843955, "deg_top50_sum": 80, "deg_top50_sum_covered_genes": 80, "per_type_logfc_r": {"T cell": 0.7028, "B cell": 0.4091, "Monocyte": 0.8035, "NK cell": 0.7244}}, "lora": {"comp_l1": 0.478123841402283, "logfc_r_mean_cp10k": 0.7798976158583719, "deg_top50_sum": 116, "partial_r_mean_cp10k": 0.6572469324283057, "coverage": 0.7569882352941176}, "broadcast": {"logfc_r_mean_cp10k": 0.6675205927010471, "logfc_r_mean_cp10k_covered_genes": 0.6645054938918955, "deg_top50_sum": 71}}
|
| 70 |
+
{"dataset": "imperial_eskd", "comparison": "fewshot", "draw": 1, "bayesprism": {"comp_l1": 1.2602294000362035, "comp_l1_by_variant": {"final_cells": 1.2602294000362035, "first_cells": 1.1404018882964686, "final_mrna": 1.2599940346426761, "first_mrna": 1.112205833851178}, "logfc_r_mean_cp10k": 0.6313284983048936, "partial_r_mean_cp10k": 0.27947517509494135, "logfc_r_mean_cp10k_covered_genes": 0.6348115703540999, "deg_top50_sum": 96, "deg_top50_sum_covered_genes": 96, "per_type_logfc_r": {"T cell": 0.684, "B cell": 0.3679, "Monocyte": 0.7474, "NK cell": 0.7259}}, "lora": {"comp_l1": 0.5160151799648663, "logfc_r_mean_cp10k": 0.7570901930571807, "deg_top50_sum": 119, "partial_r_mean_cp10k": 0.6045509566158934, "coverage": 0.7829399392381398}, "broadcast": {"logfc_r_mean_cp10k": 0.6611787175621116, "logfc_r_mean_cp10k_covered_genes": 0.6581559719140895, "deg_top50_sum": 86}}
|
| 71 |
+
{"dataset": "imperial_eskd", "comparison": "fewshot", "draw": 2, "bayesprism": {"comp_l1": 1.2822045893229197, "comp_l1_by_variant": {"final_cells": 1.2822045893229197, "first_cells": 1.1995334458513318, "final_mrna": 1.2714390668849833, "first_mrna": 1.1631974924575916}, "logfc_r_mean_cp10k": 0.6399604252827782, "partial_r_mean_cp10k": 0.3255231703400274, "logfc_r_mean_cp10k_covered_genes": 0.64365234598455, "deg_top50_sum": 83, "deg_top50_sum_covered_genes": 83, "per_type_logfc_r": {"T cell": 0.695, "B cell": 0.4238, "Monocyte": 0.7206, "NK cell": 0.7205}}, "lora": {"comp_l1": 0.5117644876225123, "logfc_r_mean_cp10k": 0.7139546009415169, "deg_top50_sum": 103, "partial_r_mean_cp10k": 0.5764372157359006, "coverage": 0.6650339110776187}, "broadcast": {"logfc_r_mean_cp10k": 0.6433613217171205, "logfc_r_mean_cp10k_covered_genes": 0.6404865393732804, "deg_top50_sum": 65}}
|
| 72 |
+
{"dataset": "imperial_eskd", "comparison": "fewshot", "draw": 3, "bayesprism": {"comp_l1": 1.2606978017175978, "comp_l1_by_variant": {"final_cells": 1.2606978017175978, "first_cells": 1.173508523665221, "final_mrna": 1.2547269731637012, "first_mrna": 1.1465540987906442}, "logfc_r_mean_cp10k": 0.6273182344451201, "partial_r_mean_cp10k": 0.28547510203613663, "logfc_r_mean_cp10k_covered_genes": 0.6312081095100865, "deg_top50_sum": 69, "deg_top50_sum_covered_genes": 69, "per_type_logfc_r": {"T cell": 0.665, "B cell": 0.3558, "Monocyte": 0.7391, "NK cell": 0.7494}}, "lora": {"comp_l1": 0.5275199844470723, "logfc_r_mean_cp10k": 0.7455905986374256, "deg_top50_sum": 117, "partial_r_mean_cp10k": 0.6007400715066787, "coverage": 0.842680564523568}, "broadcast": {"logfc_r_mean_cp10k": 0.6538959233815048, "logfc_r_mean_cp10k_covered_genes": 0.6513541840409236, "deg_top50_sum": 76}}
|
| 73 |
+
{"dataset": "imperial_eskd", "comparison": "fewshot", "draw": 4, "bayesprism": {"comp_l1": 1.2836498121844566, "comp_l1_by_variant": {"final_cells": 1.2836498121844566, "first_cells": 1.1942672475358052, "final_mrna": 1.282521324443105, "first_mrna": 1.1775761658792774}, "logfc_r_mean_cp10k": 0.5938215544028191, "partial_r_mean_cp10k": 0.2437258990135912, "logfc_r_mean_cp10k_covered_genes": 0.59799600379165, "deg_top50_sum": 96, "deg_top50_sum_covered_genes": 96, "per_type_logfc_r": {"T cell": 0.5616, "B cell": 0.342, "Monocyte": 0.7435, "NK cell": 0.7282}}, "lora": {"comp_l1": 0.48142813711177446, "logfc_r_mean_cp10k": 0.7492433446082651, "deg_top50_sum": 113, "partial_r_mean_cp10k": 0.575230848198632, "coverage": 0.6369376110747358}, "broadcast": {"logfc_r_mean_cp10k": 0.6565426449733607, "logfc_r_mean_cp10k_covered_genes": 0.6528453910767952, "deg_top50_sum": 76}}
|
| 74 |
+
condition NPZ not found: /home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/generated/imperial_eskd/pooled/conditions_pooled_raw.npz (pass --condition ARM=PATH or --no-contrast)
|
| 75 |
+
2026-10-06T18:49:43 WARN: imperial_eskd cv
|
| 76 |
+
{"dataset": "giroux2022", "comparison": "fewshot", "draw": 0, "bayesprism": {"comp_l1": 1.8027415837285268, "comp_l1_by_variant": {"final_cells": 1.8027415837285268, "first_cells": 1.3172890365195966, "final_mrna": 1.5952789694759482, "first_mrna": 1.0142832141087779}, "logfc_r_mean_cp10k": 0.42403489559228, "partial_r_mean_cp10k": 0.08333079676882724, "logfc_r_mean_cp10k_covered_genes": 0.4244868945506697, "deg_top50_sum": 0, "deg_top50_sum_covered_genes": 0, "per_type_logfc_r": {"T cell": 0.4132, "B cell": 0.4211, "Monocyte": 0.5517, "NK cell": 0.3101}}, "lora": {"comp_l1": 0.27193277679338645, "logfc_r_mean_cp10k": 0.243392212681724, "deg_top50_sum": 0, "partial_r_mean_cp10k": 0.15035969319388034, "coverage": 0.96}, "broadcast": {"logfc_r_mean_cp10k": 0.4709654147613206, "logfc_r_mean_cp10k_covered_genes": 0.47465918591696865, "deg_top50_sum": 0}}
|
| 77 |
+
{"dataset": "giroux2022", "comparison": "fewshot", "draw": 1, "bayesprism": {"comp_l1": 1.7599169676086206, "comp_l1_by_variant": {"final_cells": 1.7599169676086206, "first_cells": 0.7385246862506405, "final_mrna": 1.5255549689756318, "first_mrna": 0.4823111070328001}, "logfc_r_mean_cp10k": 0.4000648137127926, "partial_r_mean_cp10k": 0.09411226085159383, "logfc_r_mean_cp10k_covered_genes": 0.4028078113970014, "deg_top50_sum": 0, "deg_top50_sum_covered_genes": 0, "per_type_logfc_r": {"T cell": 0.4506, "B cell": 0.1739, "Monocyte": 0.5981, "NK cell": 0.3777}}, "lora": {"comp_l1": 0.22320689222847964, "logfc_r_mean_cp10k": 0.35634959703435354, "deg_top50_sum": 1, "partial_r_mean_cp10k": 0.2523250741286013, "coverage": 0.9526666666666667}, "broadcast": {"logfc_r_mean_cp10k": 0.5012527613870964, "logfc_r_mean_cp10k_covered_genes": 0.5013362659252331, "deg_top50_sum": 0}}
|
| 78 |
+
{"dataset": "giroux2022", "comparison": "fewshot", "draw": 2, "bayesprism": {"comp_l1": 1.6999423714160045, "comp_l1_by_variant": {"final_cells": 1.6999423714160045, "first_cells": 0.8243399278333422, "final_mrna": 1.4183639835960766, "first_mrna": 0.5085881594931704}, "logfc_r_mean_cp10k": 0.409140060748368, "partial_r_mean_cp10k": 0.12680783898552012, "logfc_r_mean_cp10k_covered_genes": 0.4096064238113356, "deg_top50_sum": 0, "deg_top50_sum_covered_genes": 0, "per_type_logfc_r": {"T cell": 0.2952, "B cell": 0.4492, "Monocyte": 0.6047, "NK cell": 0.2875}}, "lora": {"comp_l1": 0.2907494763185344, "logfc_r_mean_cp10k": 0.230513969445487, "deg_top50_sum": 0, "partial_r_mean_cp10k": 0.15040912208962715, "coverage": 0.9533333333333334}, "broadcast": {"logfc_r_mean_cp10k": 0.39643909018360535, "logfc_r_mean_cp10k_covered_genes": 0.3992349402913996, "deg_top50_sum": 0}}
|
| 79 |
+
{"dataset": "giroux2022", "comparison": "fewshot", "draw": 3, "bayesprism": {"comp_l1": 1.607186467641661, "comp_l1_by_variant": {"final_cells": 1.607186467641661, "first_cells": 0.6222394921050891, "final_mrna": 1.3159875062686455, "first_mrna": 0.4005043856076437}, "logfc_r_mean_cp10k": 0.4986149233422867, "partial_r_mean_cp10k": 0.20182387848262553, "logfc_r_mean_cp10k_covered_genes": 0.49903850451226733, "deg_top50_sum": 0, "deg_top50_sum_covered_genes": 0, "per_type_logfc_r": {"T cell": 0.4518, "B cell": 0.5125, "Monocyte": 0.6107, "NK cell": 0.4194}}, "lora": {"comp_l1": 0.2688359990061283, "logfc_r_mean_cp10k": 0.24186295766157903, "deg_top50_sum": 0, "partial_r_mean_cp10k": 0.14697551642627654, "coverage": 0.9624242424242424}, "broadcast": {"logfc_r_mean_cp10k": 0.4657480400911531, "logfc_r_mean_cp10k_covered_genes": 0.467863802750279, "deg_top50_sum": 0}}
|
| 80 |
+
{"dataset": "giroux2022", "comparison": "fewshot", "draw": 4, "bayesprism": {"comp_l1": 0.7421549655922297, "comp_l1_by_variant": {"final_cells": 0.7421549655922297, "first_cells": 0.5690064517688505, "final_mrna": 0.5021313182781816, "first_mrna": 0.3586912819884033}, "logfc_r_mean_cp10k": 0.5441509891552919, "partial_r_mean_cp10k": 0.2027735798353649, "logfc_r_mean_cp10k_covered_genes": 0.5461575303353852, "deg_top50_sum": 0, "deg_top50_sum_covered_genes": 0, "per_type_logfc_r": {"T cell": 0.5308, "B cell": 0.3848, "Monocyte": 0.7311, "NK cell": 0.53}}, "lora": {"comp_l1": 0.2274734478129418, "logfc_r_mean_cp10k": 0.41085265458568265, "deg_top50_sum": 0, "partial_r_mean_cp10k": 0.24279747694617004, "coverage": 0.9766666666666667}, "broadcast": {"logfc_r_mean_cp10k": 0.6280870820074911, "logfc_r_mean_cp10k_covered_genes": 0.6282955787542766, "deg_top50_sum": 0}}
|
| 81 |
+
{"dataset": "giroux2022", "comparison": "cv", "draw": null, "bayesprism": {"comp_l1": 1.0130932525218068, "comp_l1_by_variant": {"final_cells": 1.0130932525218068, "first_cells": 0.5495314839645408, "final_mrna": 0.734796598559527, "first_mrna": 0.3561888525958372}, "logfc_r_mean_cp10k": 0.4619774813232209, "partial_r_mean_cp10k": 0.057257053963693766, "logfc_r_mean_cp10k_covered_genes": 0.4653101345055838, "deg_top50_sum": 4, "deg_top50_sum_covered_genes": 4, "per_type_logfc_r": {"T cell": 0.3647, "B cell": 0.3919, "Monocyte": 0.6866, "NK cell": 0.4047}}, "lora": {"comp_l1": 0.2580442754044374, "logfc_r_mean_cp10k": 0.46094848304079133, "deg_top50_sum": 5, "partial_r_mean_cp10k": 0.3230347269696575, "coverage": 0.9280555555555555}, "broadcast": {"logfc_r_mean_cp10k": 0.5752659929360795, "logfc_r_mean_cp10k_covered_genes": 0.5757467455051679, "deg_top50_sum": 5}}
|
| 82 |
+
# BayesPrism with the same paired donors vs LoRA (paired benchmark)
|
| 83 |
+
|
| 84 |
+
Same donors, observed truth, contrast and scorer as the benchmark (score_ft.py functions; the recomputed bulk broadcast equals the stored one in every split). BayesPrism reference = the split's paired donors' observed cells (states = gate | AIFI L2); mixture = the test samples' real bulk. Composition L1 in cell units (lower is better): 'final' and 'first' BayesPrism theta. logFC r = case-vs-control per-type logFC (mean CP10K) correlation with observed, mean of T/B/Mono/NK; BayesPrism Z zero-filled on genes it lacks ('covered' = only genes it models, broadcast recomputed there). DEG = top-50 overlap summed over types.
|
| 85 |
+
|
| 86 |
+
## Five-donor draws (mean ± SD over 5 draws)
|
| 87 |
+
|
| 88 |
+
| cohort | L1 BP final | L1 BP first | L1 LoRA-5 | r BP | r BP (covered) | r LoRA-5 | r broadcast | partial r BP | partial r LoRA-5 | DEG BP | DEG LoRA-5 | DEG broadcast |
|
| 89 |
+
|---|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|
|
| 90 |
+
| Sepsis (5) | 1.177 ± 0.038 | 0.838 ± 0.158 | 0.466 ± 0.044 | 0.259 ± 0.043 | 0.270 ± 0.045 | 0.350 ± 0.162 | 0.320 ± 0.024 | 0.129 ± 0.035 | 0.305 ± 0.189 | 29.0 ± 6.2 | 16.4 ± 6.7 | 20.0 ± 1.9 |
|
| 91 |
+
| Ji ChILI (5) | 1.085 ± 0.024 | 1.040 ± 0.031 | 0.513 ± 0.115 | 0.207 ± 0.015 | 0.210 ± 0.015 | 0.333 ± 0.077 | 0.234 ± 0.026 | 0.042 ± 0.020 | 0.300 ± 0.076 | 1.6 ± 1.7 | 0.2 ± 0.4 | 1.0 ± 2.2 |
|
| 92 |
+
| Imperial (5) | 1.273 ± 0.012 | 1.183 ± 0.027 | 0.503 ± 0.022 | 0.630 ± 0.024 | 0.634 ± 0.024 | 0.749 ± 0.024 | 0.656 ± 0.009 | 0.292 ± 0.035 | 0.603 ± 0.033 | 84.8 ± 11.5 | 113.6 ± 6.3 | 74.8 ± 7.7 |
|
| 93 |
+
| Giroux (5) | 1.522 ± 0.442 | 0.814 ± 0.298 | 0.256 ± 0.030 | 0.455 ± 0.063 | 0.456 ± 0.063 | 0.297 ± 0.082 | 0.492 ± 0.085 | 0.142 ± 0.058 | 0.189 ± 0.054 | 0.0 ± 0.0 | 0.2 ± 0.4 | 0.0 ± 0.0 |
|
| 94 |
+
|
| 95 |
+
Draws where LoRA-5 beats BayesPrism (all cohorts, 20 draws): composition 20 (vs BayesPrism's better theta), logFC r 14, partial r 18, top-50 DEG 7 (ties 6).
|
| 96 |
+
|
| 97 |
+
## 5-fold CV, pooled (BayesPrism reference = each fold's LoRA training + inner-validation pairs)
|
| 98 |
+
|
| 99 |
+
| cohort | L1 BP final | L1 BP first | L1 LoRA | L1 scSemi | r BP | r BP (covered) | r LoRA | r scSemi | r broadcast | partial r BP | partial r LoRA | DEG BP | DEG LoRA | DEG scSemi | DEG broadcast |
|
| 100 |
+
|---|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|
|
| 101 |
+
| Ji ChILI | 1.098 | 1.065 | 0.504 | 0.575 | 0.241 | 0.251 | 0.479 | 0.273 | 0.240 | 0.077 | 0.452 | 4 | 15 | 0 | 4 |
|
| 102 |
+
| Giroux | 1.013 | 0.550 | 0.258 | – | 0.462 | 0.465 | 0.461 | – | 0.575 | 0.057 | 0.323 | 4 | 5 | – | 5 |
|
| 103 |
+
2026-10-06T18:49:51 finalize done; failed=8
|
paper-releases/pbisc-diffusion/updates/20261008T022410Z-deconv-scsemi-deck/diffusion/dataset/server_1/model8/plans/bayesprism_paired_20261006/logs/finalize_cx_20261008T105628.log
ADDED
|
@@ -0,0 +1,126 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"dataset": "immicro", "comparison": "fewshot", "draw": 0, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.5506126156345467, "comp_l1_by_variant": {"final_cells": 0.5506126156345467, "first_cells": 0.5506126156345467, "final_mrna": 0.63155753806753, "first_mrna": 0.63155753806753}, "mean_fraction": {"final_cells": {"T cell": 0.4298, "B cell": 0.0838, "Monocyte": 0.2848, "NK cell": 0.051, "DC": 0.0254, "Other": 0.1251}, "first_cells": {"T cell": 0.4298, "B cell": 0.0838, "Monocyte": 0.2848, "NK cell": 0.051, "DC": 0.0254, "Other": 0.1251}, "final_mrna": {"T cell": 0.3984, "B cell": 0.0725, "Monocyte": 0.324, "NK cell": 0.0359, "DC": 0.0578, "Other": 0.1115}, "first_mrna": {"T cell": 0.3984, "B cell": 0.0725, "Monocyte": 0.324, "NK cell": 0.0359, "DC": 0.0578, "Other": 0.1115}}}, "lora": {"comp_l1": 0.18149050759378063, "coverage": 0.8003921568627451}, "broadcast": {}}
|
| 2 |
+
{"dataset": "immicro", "comparison": "fewshot", "draw": 1, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.4617486814629545, "comp_l1_by_variant": {"final_cells": 0.4617486814629545, "first_cells": 0.4617486814629545, "final_mrna": 0.6101145215651299, "first_mrna": 0.6101145215651299}, "mean_fraction": {"final_cells": {"T cell": 0.465, "B cell": 0.0859, "Monocyte": 0.2225, "NK cell": 0.0717, "DC": 0.0268, "Other": 0.1281}, "first_cells": {"T cell": 0.465, "B cell": 0.0859, "Monocyte": 0.2225, "NK cell": 0.0717, "DC": 0.0268, "Other": 0.1281}, "final_mrna": {"T cell": 0.3976, "B cell": 0.077, "Monocyte": 0.2644, "NK cell": 0.0508, "DC": 0.062, "Other": 0.1482}, "first_mrna": {"T cell": 0.3976, "B cell": 0.077, "Monocyte": 0.2644, "NK cell": 0.0508, "DC": 0.062, "Other": 0.1482}}}, "lora": {"comp_l1": 0.1566780422080866, "coverage": 0.8619607843137255}, "broadcast": {}}
|
| 3 |
+
{"dataset": "immicro", "comparison": "fewshot", "draw": 2, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.49552826442174286, "comp_l1_by_variant": {"final_cells": 0.49552826442174286, "first_cells": 0.49552826442174286, "final_mrna": 0.6257486040187901, "first_mrna": 0.6257486040187901}, "mean_fraction": {"final_cells": {"T cell": 0.4404, "B cell": 0.1075, "Monocyte": 0.2886, "NK cell": 0.078, "DC": 0.017, "Other": 0.0684}, "first_cells": {"T cell": 0.4404, "B cell": 0.1075, "Monocyte": 0.2886, "NK cell": 0.078, "DC": 0.017, "Other": 0.0684}, "final_mrna": {"T cell": 0.3813, "B cell": 0.0965, "Monocyte": 0.3396, "NK cell": 0.0589, "DC": 0.0403, "Other": 0.0833}, "first_mrna": {"T cell": 0.3813, "B cell": 0.0965, "Monocyte": 0.3396, "NK cell": 0.0589, "DC": 0.0403, "Other": 0.0833}}}, "lora": {"comp_l1": 0.17248763359752534, "coverage": 0.7860784313725491}, "broadcast": {}}
|
| 4 |
+
{"dataset": "immicro", "comparison": "fewshot", "draw": 3, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.3500687319968211, "comp_l1_by_variant": {"final_cells": 0.3500687319968211, "first_cells": 0.3500687319968211, "final_mrna": 0.4609720856629254, "first_mrna": 0.4609720856629254}, "mean_fraction": {"final_cells": {"T cell": 0.5061, "B cell": 0.0907, "Monocyte": 0.2668, "NK cell": 0.0961, "DC": 0.026, "Other": 0.0143}, "first_cells": {"T cell": 0.5061, "B cell": 0.0907, "Monocyte": 0.2668, "NK cell": 0.0961, "DC": 0.026, "Other": 0.0143}, "final_mrna": {"T cell": 0.4578, "B cell": 0.0846, "Monocyte": 0.3093, "NK cell": 0.0682, "DC": 0.0598, "Other": 0.0204}, "first_mrna": {"T cell": 0.4578, "B cell": 0.0846, "Monocyte": 0.3093, "NK cell": 0.0682, "DC": 0.0598, "Other": 0.0204}}}, "lora": {"comp_l1": 0.18654520396343544, "coverage": 0.7829411764705883}, "broadcast": {}}
|
| 5 |
+
{"dataset": "immicro", "comparison": "fewshot", "draw": 4, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.599083099320475, "comp_l1_by_variant": {"final_cells": 0.599083099320475, "first_cells": 0.599083099320475, "final_mrna": 0.7165605084426594, "first_mrna": 0.7165605084426594}, "mean_fraction": {"final_cells": {"T cell": 0.4112, "B cell": 0.0791, "Monocyte": 0.2872, "NK cell": 0.0522, "DC": 0.0217, "Other": 0.1486}, "first_cells": {"T cell": 0.4112, "B cell": 0.0791, "Monocyte": 0.2872, "NK cell": 0.0522, "DC": 0.0217, "Other": 0.1486}, "final_mrna": {"T cell": 0.36, "B cell": 0.071, "Monocyte": 0.3152, "NK cell": 0.0373, "DC": 0.0452, "Other": 0.1714}, "first_mrna": {"T cell": 0.36, "B cell": 0.071, "Monocyte": 0.3152, "NK cell": 0.0373, "DC": 0.0452, "Other": 0.1714}}}, "lora": {"comp_l1": 0.18059655915509124, "coverage": 0.7309803921568627}, "broadcast": {}}
|
| 6 |
+
{"dataset": "immicro", "comparison": "cv", "draw": null, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.5065033643252109, "comp_l1_by_variant": {"final_cells": 0.5065033643252109, "first_cells": 0.5065033643252109, "final_mrna": 0.6252519106334087, "first_mrna": 0.6252519106334087}, "mean_fraction": {"final_cells": {"T cell": 0.4656, "B cell": 0.0984, "Monocyte": 0.2073, "NK cell": 0.0362, "DC": 0.0211, "Other": 0.1715}, "first_cells": {"T cell": 0.4656, "B cell": 0.0984, "Monocyte": 0.2073, "NK cell": 0.0362, "DC": 0.0211, "Other": 0.1715}, "final_mrna": {"T cell": 0.4082, "B cell": 0.0879, "Monocyte": 0.2404, "NK cell": 0.0264, "DC": 0.0479, "Other": 0.1893}, "first_mrna": {"T cell": 0.4082, "B cell": 0.0879, "Monocyte": 0.2404, "NK cell": 0.0264, "DC": 0.0479, "Other": 0.1893}}}, "lora": {"comp_l1": 0.16172890350786212, "coverage": 0.7842990654205607}, "broadcast": {}}
|
| 7 |
+
{"dataset": "ye2026_sepsis", "comparison": "fewshot", "draw": 0, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.594921646565701, "comp_l1_by_variant": {"final_cells": 0.594921646565701, "first_cells": 0.594921646565701, "final_mrna": 0.6285146093050138, "first_mrna": 0.6285146093050138}, "mean_fraction": {"final_cells": {"T cell": 0.3685, "B cell": 0.1149, "Monocyte": 0.1853, "NK cell": 0.1496, "DC": 0.0709, "Other": 0.1109}, "first_cells": {"T cell": 0.3685, "B cell": 0.1149, "Monocyte": 0.1853, "NK cell": 0.1496, "DC": 0.0709, "Other": 0.1109}, "final_mrna": {"T cell": 0.3379, "B cell": 0.1255, "Monocyte": 0.2116, "NK cell": 0.1248, "DC": 0.1471, "Other": 0.0532}, "first_mrna": {"T cell": 0.3379, "B cell": 0.1255, "Monocyte": 0.2116, "NK cell": 0.1248, "DC": 0.1471, "Other": 0.0532}}, "logfc_r_mean_cp10k": 0.07598737573212785, "partial_r_mean_cp10k": 0.008099941756831479, "logfc_r_mean_cp10k_covered_genes": 0.07598737573212785, "deg_top50_sum": 22, "deg_top50_sum_covered_genes": 22, "per_type_logfc_r": {"T cell": -0.0351, "B cell": 0.0345, "Monocyte": 0.2635, "NK cell": 0.0411}}, "lora": {"comp_l1": 0.4723364713369038, "coverage": 0.7058874458874459, "logfc_r_mean_cp10k": 0.4359905652333528, "deg_top50_sum": 7, "partial_r_mean_cp10k": 0.4060106147529024}, "broadcast": {"logfc_r_mean_cp10k": 0.3359901153517466, "logfc_r_mean_cp10k_covered_genes": 0.3359901153517466, "deg_top50_sum": 23}}
|
| 8 |
+
{"dataset": "ye2026_sepsis", "comparison": "fewshot", "draw": 1, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.5458984505842515, "comp_l1_by_variant": {"final_cells": 0.5458984505842515, "first_cells": 0.5458984505842515, "final_mrna": 0.5639663397037358, "first_mrna": 0.5639663397037358}, "mean_fraction": {"final_cells": {"T cell": 0.3869, "B cell": 0.127, "Monocyte": 0.2163, "NK cell": 0.1444, "DC": 0.0356, "Other": 0.0898}, "first_cells": {"T cell": 0.3869, "B cell": 0.127, "Monocyte": 0.2163, "NK cell": 0.1444, "DC": 0.0356, "Other": 0.0898}, "final_mrna": {"T cell": 0.3603, "B cell": 0.1048, "Monocyte": 0.2872, "NK cell": 0.1132, "DC": 0.0734, "Other": 0.061}, "first_mrna": {"T cell": 0.3603, "B cell": 0.1048, "Monocyte": 0.2872, "NK cell": 0.1132, "DC": 0.0734, "Other": 0.061}}, "logfc_r_mean_cp10k": 0.0506459348741054, "partial_r_mean_cp10k": -0.02049189218265121, "logfc_r_mean_cp10k_covered_genes": 0.0506459348741054, "deg_top50_sum": 26, "deg_top50_sum_covered_genes": 26, "per_type_logfc_r": {"T cell": 0.0867, "B cell": -0.1838, "Monocyte": 0.3041, "NK cell": -0.0043}}, "lora": {"comp_l1": 0.4725655492172997, "coverage": 0.6619047619047619, "logfc_r_mean_cp10k": 0.08123703391754132, "deg_top50_sum": 16, "partial_r_mean_cp10k": -0.018935679613692524}, "broadcast": {"logfc_r_mean_cp10k": 0.27941436238089834, "logfc_r_mean_cp10k_covered_genes": 0.27941436238089834, "deg_top50_sum": 20}}
|
| 9 |
+
{"dataset": "ye2026_sepsis", "comparison": "fewshot", "draw": 2, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.5682407267081256, "comp_l1_by_variant": {"final_cells": 0.5682407267081256, "first_cells": 0.5682407267081256, "final_mrna": 0.568766565188311, "first_mrna": 0.568766565188311}, "mean_fraction": {"final_cells": {"T cell": 0.3554, "B cell": 0.1283, "Monocyte": 0.2051, "NK cell": 0.1588, "DC": 0.0374, "Other": 0.1149}, "first_cells": {"T cell": 0.3554, "B cell": 0.1283, "Monocyte": 0.2051, "NK cell": 0.1588, "DC": 0.0374, "Other": 0.1149}, "final_mrna": {"T cell": 0.3466, "B cell": 0.1104, "Monocyte": 0.2715, "NK cell": 0.1301, "DC": 0.0838, "Other": 0.0576}, "first_mrna": {"T cell": 0.3466, "B cell": 0.1104, "Monocyte": 0.2715, "NK cell": 0.1301, "DC": 0.0838, "Other": 0.0576}}, "logfc_r_mean_cp10k": 0.069436100894467, "partial_r_mean_cp10k": -0.03391247227739959, "logfc_r_mean_cp10k_covered_genes": 0.069436100894467, "deg_top50_sum": 25, "deg_top50_sum_covered_genes": 25, "per_type_logfc_r": {"T cell": 0.0311, "B cell": -0.1007, "Monocyte": 0.3167, "NK cell": 0.0306}}, "lora": {"comp_l1": 0.4986968896699433, "coverage": 0.7606060606060606, "logfc_r_mean_cp10k": 0.32721881191208096, "deg_top50_sum": 26, "partial_r_mean_cp10k": 0.31447561738686153}, "broadcast": {"logfc_r_mean_cp10k": 0.3386821732452143, "logfc_r_mean_cp10k_covered_genes": 0.3386821732452143, "deg_top50_sum": 18}}
|
| 10 |
+
{"dataset": "ye2026_sepsis", "comparison": "fewshot", "draw": 3, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.6003763342155404, "comp_l1_by_variant": {"final_cells": 0.6003763342155404, "first_cells": 0.6003763342155404, "final_mrna": 0.5956927069312322, "first_mrna": 0.5956927069312322}, "mean_fraction": {"final_cells": {"T cell": 0.426, "B cell": 0.0933, "Monocyte": 0.1617, "NK cell": 0.1723, "DC": 0.0436, "Other": 0.1031}, "first_cells": {"T cell": 0.426, "B cell": 0.0933, "Monocyte": 0.1617, "NK cell": 0.1723, "DC": 0.0436, "Other": 0.1031}, "final_mrna": {"T cell": 0.3633, "B cell": 0.1204, "Monocyte": 0.2045, "NK cell": 0.1299, "DC": 0.1106, "Other": 0.0712}, "first_mrna": {"T cell": 0.3633, "B cell": 0.1204, "Monocyte": 0.2045, "NK cell": 0.1299, "DC": 0.1106, "Other": 0.0712}}, "logfc_r_mean_cp10k": 0.10319244044179009, "partial_r_mean_cp10k": 0.025522354245958428, "logfc_r_mean_cp10k_covered_genes": 0.10319244044179009, "deg_top50_sum": 21, "deg_top50_sum_covered_genes": 21, "per_type_logfc_r": {"T cell": -0.0267, "B cell": 0.088, "Monocyte": 0.2895, "NK cell": 0.062}}, "lora": {"comp_l1": 0.4968788955478774, "coverage": 0.6553246753246753, "logfc_r_mean_cp10k": 0.4108841123350609, "deg_top50_sum": 17, "partial_r_mean_cp10k": 0.3622128751991822}, "broadcast": {"logfc_r_mean_cp10k": 0.3264466405549955, "logfc_r_mean_cp10k_covered_genes": 0.3264466405549955, "deg_top50_sum": 19}}
|
| 11 |
+
{"dataset": "ye2026_sepsis", "comparison": "fewshot", "draw": 4, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.6015123278200128, "comp_l1_by_variant": {"final_cells": 0.6015123278200128, "first_cells": 0.6015123278200128, "final_mrna": 0.5871885532406123, "first_mrna": 0.5871885532406123}, "mean_fraction": {"final_cells": {"T cell": 0.4015, "B cell": 0.1026, "Monocyte": 0.1537, "NK cell": 0.1146, "DC": 0.0517, "Other": 0.1759}, "first_cells": {"T cell": 0.4015, "B cell": 0.1026, "Monocyte": 0.1537, "NK cell": 0.1146, "DC": 0.0517, "Other": 0.1759}, "final_mrna": {"T cell": 0.3823, "B cell": 0.1039, "Monocyte": 0.199, "NK cell": 0.0968, "DC": 0.1334, "Other": 0.0846}, "first_mrna": {"T cell": 0.3823, "B cell": 0.1039, "Monocyte": 0.199, "NK cell": 0.0968, "DC": 0.1334, "Other": 0.0846}}, "logfc_r_mean_cp10k": 0.04659377114585425, "partial_r_mean_cp10k": -0.028782733910565704, "logfc_r_mean_cp10k_covered_genes": 0.04659377114585425, "deg_top50_sum": 22, "deg_top50_sum_covered_genes": 22, "per_type_logfc_r": {"T cell": 0.0341, "B cell": -0.1489, "Monocyte": 0.2647, "NK cell": 0.0365}}, "lora": {"comp_l1": 0.39049012270725986, "coverage": 0.687012987012987, "logfc_r_mean_cp10k": 0.49526136963540107, "deg_top50_sum": 16, "partial_r_mean_cp10k": 0.4597246910197378}, "broadcast": {"logfc_r_mean_cp10k": 0.3195942137906433, "logfc_r_mean_cp10k_covered_genes": 0.3195942137906433, "deg_top50_sum": 20}}
|
| 12 |
+
{"dataset": "ye2026_sepsis", "comparison": "cv", "draw": null, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.607149641152929, "comp_l1_by_variant": {"final_cells": 0.607149641152929, "first_cells": 0.607149641152929, "final_mrna": 0.6133993055664033, "first_mrna": 0.6133993055664033}, "mean_fraction": {"final_cells": {"T cell": 0.4368, "B cell": 0.1095, "Monocyte": 0.1245, "NK cell": 0.1057, "DC": 0.0794, "Other": 0.1441}, "first_cells": {"T cell": 0.4368, "B cell": 0.1095, "Monocyte": 0.1245, "NK cell": 0.1057, "DC": 0.0794, "Other": 0.1441}, "final_mrna": {"T cell": 0.4004, "B cell": 0.1125, "Monocyte": 0.1573, "NK cell": 0.0838, "DC": 0.1661, "Other": 0.0798}, "first_mrna": {"T cell": 0.4004, "B cell": 0.1125, "Monocyte": 0.1573, "NK cell": 0.0838, "DC": 0.1661, "Other": 0.0798}}, "logfc_r_mean_cp10k": -0.04533916518496376, "partial_r_mean_cp10k": -0.08141292086114983, "logfc_r_mean_cp10k_covered_genes": -0.04533916518496376, "deg_top50_sum": 0, "deg_top50_sum_covered_genes": 0, "per_type_logfc_r": {"T cell": -0.0669, "B cell": -0.1331, "Monocyte": 0.033, "NK cell": -0.0143}}, "lora": {"comp_l1": 0.3618246783833512, "coverage": 0.829349593495935, "logfc_r_mean_cp10k": 0.5267612711229187, "deg_top50_sum": 23, "partial_r_mean_cp10k": 0.46874787884206276, "deg_source": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/scores/ye2026_sepsis/pooled_deg/summary.json"}, "broadcast": {"logfc_r_mean_cp10k": 0.31594581626202867, "logfc_r_mean_cp10k_covered_genes": 0.31594581626202867, "deg_top50_sum": 21}}
|
| 13 |
+
/home/sj_server_1/miniforge3/envs/cxg_corpus/lib/python3.11/site-packages/scipy/stats/_axis_nan_policy.py:635: RuntimeWarning: Precision loss occurred in moment calculation due to catastrophic cancellation. This occurs when the data are nearly identical. Results may be unreliable.
|
| 14 |
+
res = hypotest_fun_out(*samples, axis=axis, **kwds)
|
| 15 |
+
/home/sj_server_1/miniforge3/envs/cxg_corpus/lib/python3.11/site-packages/scipy/stats/_axis_nan_policy.py:635: RuntimeWarning: Precision loss occurred in moment calculation due to catastrophic cancellation. This occurs when the data are nearly identical. Results may be unreliable.
|
| 16 |
+
res = hypotest_fun_out(*samples, axis=axis, **kwds)
|
| 17 |
+
{"dataset": "ji2025_chili", "comparison": "fewshot", "draw": 0, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 1.0251969201714595, "comp_l1_by_variant": {"final_cells": 1.0251969201714595, "first_cells": 1.0251969201714595, "final_mrna": 1.0722077602978095, "first_mrna": 1.0722077602978095}, "mean_fraction": {"final_cells": {"T cell": 0.1733, "B cell": 0.1067, "Monocyte": 0.1693, "NK cell": 0.0948, "DC": 0.0327, "Other": 0.4231}, "first_cells": {"T cell": 0.1733, "B cell": 0.1067, "Monocyte": 0.1693, "NK cell": 0.0948, "DC": 0.0327, "Other": 0.4231}, "final_mrna": {"T cell": 0.1532, "B cell": 0.1069, "Monocyte": 0.1978, "NK cell": 0.0744, "DC": 0.0512, "Other": 0.4165}, "first_mrna": {"T cell": 0.1532, "B cell": 0.1069, "Monocyte": 0.1978, "NK cell": 0.0744, "DC": 0.0512, "Other": 0.4165}}, "logfc_r_mean_cp10k": 0.18186290248209486, "partial_r_mean_cp10k": 0.00811262978595458, "logfc_r_mean_cp10k_covered_genes": 0.18186290248209486, "deg_top50_sum": 0, "deg_top50_sum_covered_genes": 0, "per_type_logfc_r": {"T cell": NaN, "B cell": NaN, "Monocyte": 0.1819, "NK cell": NaN}}, "lora": {"comp_l1": 0.46942970758493824, "coverage": 0.6992, "logfc_r_mean_cp10k": 0.38697985726899736, "deg_top50_sum": 0, "partial_r_mean_cp10k": 0.36588903106531123}, "broadcast": {"logfc_r_mean_cp10k": 0.21311861411338662, "logfc_r_mean_cp10k_covered_genes": 0.21311861411338662, "deg_top50_sum": 0}}
|
| 18 |
+
/home/sj_server_1/miniforge3/envs/cxg_corpus/lib/python3.11/site-packages/scipy/stats/_axis_nan_policy.py:635: RuntimeWarning: Precision loss occurred in moment calculation due to catastrophic cancellation. This occurs when the data are nearly identical. Results may be unreliable.
|
| 19 |
+
res = hypotest_fun_out(*samples, axis=axis, **kwds)
|
| 20 |
+
/home/sj_server_1/miniforge3/envs/cxg_corpus/lib/python3.11/site-packages/scipy/stats/_axis_nan_policy.py:635: RuntimeWarning: Precision loss occurred in moment calculation due to catastrophic cancellation. This occurs when the data are nearly identical. Results may be unreliable.
|
| 21 |
+
res = hypotest_fun_out(*samples, axis=axis, **kwds)
|
| 22 |
+
{"dataset": "ji2025_chili", "comparison": "fewshot", "draw": 1, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.8496706341193269, "comp_l1_by_variant": {"final_cells": 0.8496706341193269, "first_cells": 0.8496706341193269, "final_mrna": 0.8443286298134811, "first_mrna": 0.8443286298134811}, "mean_fraction": {"final_cells": {"T cell": 0.2263, "B cell": 0.1188, "Monocyte": 0.1734, "NK cell": 0.1309, "DC": 0.0355, "Other": 0.3151}, "first_cells": {"T cell": 0.2263, "B cell": 0.1188, "Monocyte": 0.1734, "NK cell": 0.1309, "DC": 0.0355, "Other": 0.3151}, "final_mrna": {"T cell": 0.2135, "B cell": 0.1092, "Monocyte": 0.2086, "NK cell": 0.1187, "DC": 0.0698, "Other": 0.2802}, "first_mrna": {"T cell": 0.2135, "B cell": 0.1092, "Monocyte": 0.2086, "NK cell": 0.1187, "DC": 0.0698, "Other": 0.2802}}, "logfc_r_mean_cp10k": 0.18422916821886148, "partial_r_mean_cp10k": -0.03580575846082884, "logfc_r_mean_cp10k_covered_genes": 0.18422916821886148, "deg_top50_sum": 0, "deg_top50_sum_covered_genes": 0, "per_type_logfc_r": {"T cell": NaN, "B cell": NaN, "Monocyte": 0.1842, "NK cell": NaN}}, "lora": {"comp_l1": 0.3891842501829222, "coverage": 0.8551282051282051, "logfc_r_mean_cp10k": 0.27568480061697936, "deg_top50_sum": 1, "partial_r_mean_cp10k": 0.2420147625491504}, "broadcast": {"logfc_r_mean_cp10k": 0.20791984755382747, "logfc_r_mean_cp10k_covered_genes": 0.20791984755382747, "deg_top50_sum": 0}}
|
| 23 |
+
/home/sj_server_1/miniforge3/envs/cxg_corpus/lib/python3.11/site-packages/scipy/stats/_axis_nan_policy.py:635: RuntimeWarning: Precision loss occurred in moment calculation due to catastrophic cancellation. This occurs when the data are nearly identical. Results may be unreliable.
|
| 24 |
+
res = hypotest_fun_out(*samples, axis=axis, **kwds)
|
| 25 |
+
/home/sj_server_1/miniforge3/envs/cxg_corpus/lib/python3.11/site-packages/scipy/stats/_axis_nan_policy.py:635: RuntimeWarning: Precision loss occurred in moment calculation due to catastrophic cancellation. This occurs when the data are nearly identical. Results may be unreliable.
|
| 26 |
+
res = hypotest_fun_out(*samples, axis=axis, **kwds)
|
| 27 |
+
{"dataset": "ji2025_chili", "comparison": "fewshot", "draw": 2, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.865967063865684, "comp_l1_by_variant": {"final_cells": 0.865967063865684, "first_cells": 0.865967063865684, "final_mrna": 0.9243804613448391, "first_mrna": 0.9243804613448391}, "mean_fraction": {"final_cells": {"T cell": 0.2356, "B cell": 0.1022, "Monocyte": 0.2196, "NK cell": 0.0622, "DC": 0.0296, "Other": 0.3508}, "first_cells": {"T cell": 0.2356, "B cell": 0.1022, "Monocyte": 0.2196, "NK cell": 0.0622, "DC": 0.0296, "Other": 0.3508}, "final_mrna": {"T cell": 0.1917, "B cell": 0.0953, "Monocyte": 0.2702, "NK cell": 0.0585, "DC": 0.0576, "Other": 0.3267}, "first_mrna": {"T cell": 0.1917, "B cell": 0.0953, "Monocyte": 0.2702, "NK cell": 0.0585, "DC": 0.0576, "Other": 0.3267}}, "logfc_r_mean_cp10k": 0.07709960415700085, "partial_r_mean_cp10k": -0.010564281489781741, "logfc_r_mean_cp10k_covered_genes": 0.07709960415700085, "deg_top50_sum": 0, "deg_top50_sum_covered_genes": 0, "per_type_logfc_r": {"T cell": NaN, "B cell": NaN, "Monocyte": 0.0771, "NK cell": NaN}}, "lora": {"comp_l1": 0.5825794078861584, "coverage": 0.5297435897435897, "logfc_r_mean_cp10k": 0.28938101620694295, "deg_top50_sum": 0, "partial_r_mean_cp10k": 0.24803917247992743}, "broadcast": {"logfc_r_mean_cp10k": 0.263723888008903, "logfc_r_mean_cp10k_covered_genes": 0.263723888008903, "deg_top50_sum": 5}}
|
| 28 |
+
/home/sj_server_1/miniforge3/envs/cxg_corpus/lib/python3.11/site-packages/scipy/stats/_axis_nan_policy.py:635: RuntimeWarning: Precision loss occurred in moment calculation due to catastrophic cancellation. This occurs when the data are nearly identical. Results may be unreliable.
|
| 29 |
+
res = hypotest_fun_out(*samples, axis=axis, **kwds)
|
| 30 |
+
/home/sj_server_1/miniforge3/envs/cxg_corpus/lib/python3.11/site-packages/scipy/stats/_axis_nan_policy.py:635: RuntimeWarning: Precision loss occurred in moment calculation due to catastrophic cancellation. This occurs when the data are nearly identical. Results may be unreliable.
|
| 31 |
+
res = hypotest_fun_out(*samples, axis=axis, **kwds)
|
| 32 |
+
{"dataset": "ji2025_chili", "comparison": "fewshot", "draw": 3, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.7607882960499378, "comp_l1_by_variant": {"final_cells": 0.7607882960499378, "first_cells": 0.7607882960499378, "final_mrna": 0.8501451075273423, "first_mrna": 0.8501451075273423}, "mean_fraction": {"final_cells": {"T cell": 0.283, "B cell": 0.0993, "Monocyte": 0.1991, "NK cell": 0.0767, "DC": 0.0436, "Other": 0.2982}, "first_cells": {"T cell": 0.283, "B cell": 0.0993, "Monocyte": 0.1991, "NK cell": 0.0767, "DC": 0.0436, "Other": 0.2982}, "final_mrna": {"T cell": 0.2421, "B cell": 0.0951, "Monocyte": 0.2325, "NK cell": 0.0536, "DC": 0.0803, "Other": 0.2964}, "first_mrna": {"T cell": 0.2421, "B cell": 0.0951, "Monocyte": 0.2325, "NK cell": 0.0536, "DC": 0.0803, "Other": 0.2964}}, "logfc_r_mean_cp10k": NaN, "partial_r_mean_cp10k": 0.018967058193383825, "logfc_r_mean_cp10k_covered_genes": NaN, "deg_top50_sum": 0, "deg_top50_sum_covered_genes": 0, "per_type_logfc_r": {"T cell": NaN, "B cell": NaN, "Monocyte": NaN, "NK cell": NaN}}, "lora": {"comp_l1": 0.4487906027467767, "coverage": 0.7586666666666667, "logfc_r_mean_cp10k": 0.4410983924408845, "deg_top50_sum": 0, "partial_r_mean_cp10k": 0.39982215110012487}, "broadcast": {"logfc_r_mean_cp10k": 0.2577780039471977, "logfc_r_mean_cp10k_covered_genes": 0.2577780039471977, "deg_top50_sum": 0}}
|
| 33 |
+
/home/sj_server_1/miniforge3/envs/cxg_corpus/lib/python3.11/site-packages/scipy/stats/_axis_nan_policy.py:635: RuntimeWarning: Precision loss occurred in moment calculation due to catastrophic cancellation. This occurs when the data are nearly identical. Results may be unreliable.
|
| 34 |
+
res = hypotest_fun_out(*samples, axis=axis, **kwds)
|
| 35 |
+
/home/sj_server_1/miniforge3/envs/cxg_corpus/lib/python3.11/site-packages/scipy/stats/_axis_nan_policy.py:635: RuntimeWarning: Precision loss occurred in moment calculation due to catastrophic cancellation. This occurs when the data are nearly identical. Results may be unreliable.
|
| 36 |
+
res = hypotest_fun_out(*samples, axis=axis, **kwds)
|
| 37 |
+
{"dataset": "ji2025_chili", "comparison": "fewshot", "draw": 4, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.6321183546570056, "comp_l1_by_variant": {"final_cells": 0.6321183546570056, "first_cells": 0.6321183546570056, "final_mrna": 0.7003923657443595, "first_mrna": 0.7003923657443595}, "mean_fraction": {"final_cells": {"T cell": 0.3309, "B cell": 0.1061, "Monocyte": 0.249, "NK cell": 0.0611, "DC": 0.0428, "Other": 0.21}, "first_cells": {"T cell": 0.3309, "B cell": 0.1061, "Monocyte": 0.249, "NK cell": 0.0611, "DC": 0.0428, "Other": 0.21}, "final_mrna": {"T cell": 0.2757, "B cell": 0.0962, "Monocyte": 0.2967, "NK cell": 0.0483, "DC": 0.0863, "Other": 0.1968}, "first_mrna": {"T cell": 0.2757, "B cell": 0.0962, "Monocyte": 0.2967, "NK cell": 0.0483, "DC": 0.0863, "Other": 0.1968}}, "logfc_r_mean_cp10k": 0.18872575035861355, "partial_r_mean_cp10k": 0.04587211715918757, "logfc_r_mean_cp10k_covered_genes": 0.18872575035861355, "deg_top50_sum": 0, "deg_top50_sum_covered_genes": 0, "per_type_logfc_r": {"T cell": 0.1158, "B cell": NaN, "Monocyte": 0.2617, "NK cell": NaN}}, "lora": {"comp_l1": 0.6754540168870949, "coverage": 0.6936111111111111, "logfc_r_mean_cp10k": 0.2725492074488929, "deg_top50_sum": 0, "partial_r_mean_cp10k": 0.2467295164617603}, "broadcast": {"logfc_r_mean_cp10k": 0.2254541110690231, "logfc_r_mean_cp10k_covered_genes": 0.2254541110690231, "deg_top50_sum": 0}}
|
| 38 |
+
{"dataset": "ji2025_chili", "comparison": "cv", "draw": null, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.914073829946989, "comp_l1_by_variant": {"final_cells": 0.914073829946989, "first_cells": 0.914073829946989, "final_mrna": 0.9502242518422143, "first_mrna": 0.9502242518422143}, "mean_fraction": {"final_cells": {"T cell": 0.2995, "B cell": 0.0293, "Monocyte": 0.1191, "NK cell": 0.0932, "DC": 0.0492, "Other": 0.4096}, "first_cells": {"T cell": 0.2995, "B cell": 0.0293, "Monocyte": 0.1191, "NK cell": 0.0932, "DC": 0.0492, "Other": 0.4096}, "final_mrna": {"T cell": 0.264, "B cell": 0.0314, "Monocyte": 0.1452, "NK cell": 0.0817, "DC": 0.0925, "Other": 0.3853}, "first_mrna": {"T cell": 0.264, "B cell": 0.0314, "Monocyte": 0.1452, "NK cell": 0.0817, "DC": 0.0925, "Other": 0.3853}}}, "lora": {"comp_l1": 0.5044992874562503, "coverage": 0.8569696969696969}, "broadcast": {}}
|
| 39 |
+
{"dataset": "imperial_eskd", "comparison": "fewshot", "draw": 0, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.6236244635423837, "comp_l1_by_variant": {"final_cells": 0.6236244635423837, "first_cells": 0.6236244635423837, "final_mrna": 0.6950177678827166, "first_mrna": 0.6950177678827166}, "mean_fraction": {"final_cells": {"T cell": 0.3573, "B cell": 0.055, "Monocyte": 0.3667, "NK cell": 0.1477, "DC": 0.0271, "Other": 0.0461}, "first_cells": {"T cell": 0.3573, "B cell": 0.055, "Monocyte": 0.3667, "NK cell": 0.1477, "DC": 0.0271, "Other": 0.0461}, "final_mrna": {"T cell": 0.3301, "B cell": 0.0559, "Monocyte": 0.4051, "NK cell": 0.1119, "DC": 0.0521, "Other": 0.045}, "first_mrna": {"T cell": 0.3301, "B cell": 0.0559, "Monocyte": 0.4051, "NK cell": 0.1119, "DC": 0.0521, "Other": 0.045}}, "logfc_r_mean_cp10k": 0.2925661487517866, "partial_r_mean_cp10k": 0.07366698912947896, "logfc_r_mean_cp10k_covered_genes": 0.2925661487517866, "deg_top50_sum": 31, "deg_top50_sum_covered_genes": 31, "per_type_logfc_r": {"T cell": 0.2376, "B cell": 0.0538, "Monocyte": 0.7111, "NK cell": 0.1678}}, "lora": {"comp_l1": 0.478123841402283, "coverage": 0.7569882352941176, "logfc_r_mean_cp10k": 0.7798976158583719, "deg_top50_sum": 116, "partial_r_mean_cp10k": 0.6572469324283057}, "broadcast": {"logfc_r_mean_cp10k": 0.6675205927010471, "logfc_r_mean_cp10k_covered_genes": 0.6675205927010471, "deg_top50_sum": 71}}
|
| 40 |
+
{"dataset": "imperial_eskd", "comparison": "fewshot", "draw": 1, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.6541762446688113, "comp_l1_by_variant": {"final_cells": 0.6541762446688113, "first_cells": 0.6541762446688113, "final_mrna": 0.691286377681052, "first_mrna": 0.691286377681052}, "mean_fraction": {"final_cells": {"T cell": 0.3465, "B cell": 0.0616, "Monocyte": 0.3789, "NK cell": 0.1115, "DC": 0.046, "Other": 0.0555}, "first_cells": {"T cell": 0.3465, "B cell": 0.0616, "Monocyte": 0.3789, "NK cell": 0.1115, "DC": 0.046, "Other": 0.0555}, "final_mrna": {"T cell": 0.3377, "B cell": 0.0608, "Monocyte": 0.3639, "NK cell": 0.0931, "DC": 0.0918, "Other": 0.0527}, "first_mrna": {"T cell": 0.3377, "B cell": 0.0608, "Monocyte": 0.3639, "NK cell": 0.0931, "DC": 0.0918, "Other": 0.0527}}, "logfc_r_mean_cp10k": 0.2667405815064273, "partial_r_mean_cp10k": 0.037657088158500454, "logfc_r_mean_cp10k_covered_genes": 0.2667405815064273, "deg_top50_sum": 30, "deg_top50_sum_covered_genes": 30, "per_type_logfc_r": {"T cell": 0.2465, "B cell": 0.0835, "Monocyte": 0.6418, "NK cell": 0.0951}}, "lora": {"comp_l1": 0.5160151799648663, "coverage": 0.7829399392381398, "logfc_r_mean_cp10k": 0.7570901930571807, "deg_top50_sum": 119, "partial_r_mean_cp10k": 0.6045509566158934}, "broadcast": {"logfc_r_mean_cp10k": 0.6611787175621116, "logfc_r_mean_cp10k_covered_genes": 0.6611787175621116, "deg_top50_sum": 86}}
|
| 41 |
+
{"dataset": "imperial_eskd", "comparison": "fewshot", "draw": 2, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.6855049998932723, "comp_l1_by_variant": {"final_cells": 0.6855049998932723, "first_cells": 0.6855049998932723, "final_mrna": 0.6935421290569926, "first_mrna": 0.6935421290569926}, "mean_fraction": {"final_cells": {"T cell": 0.2637, "B cell": 0.0911, "Monocyte": 0.3414, "NK cell": 0.1934, "DC": 0.0513, "Other": 0.0591}, "first_cells": {"T cell": 0.2637, "B cell": 0.0911, "Monocyte": 0.3414, "NK cell": 0.1934, "DC": 0.0513, "Other": 0.0591}, "final_mrna": {"T cell": 0.2748, "B cell": 0.096, "Monocyte": 0.3317, "NK cell": 0.1436, "DC": 0.0863, "Other": 0.0676}, "first_mrna": {"T cell": 0.2748, "B cell": 0.096, "Monocyte": 0.3317, "NK cell": 0.1436, "DC": 0.0863, "Other": 0.0676}}, "logfc_r_mean_cp10k": 0.27728225351756514, "partial_r_mean_cp10k": 0.06071705242731967, "logfc_r_mean_cp10k_covered_genes": 0.27728225351756514, "deg_top50_sum": 27, "deg_top50_sum_covered_genes": 27, "per_type_logfc_r": {"T cell": 0.2126, "B cell": 0.118, "Monocyte": 0.6202, "NK cell": 0.1583}}, "lora": {"comp_l1": 0.5117644876225123, "coverage": 0.6650339110776187, "logfc_r_mean_cp10k": 0.7139546009415169, "deg_top50_sum": 103, "partial_r_mean_cp10k": 0.5764372157359006}, "broadcast": {"logfc_r_mean_cp10k": 0.6433613217171205, "logfc_r_mean_cp10k_covered_genes": 0.6433613217171205, "deg_top50_sum": 65}}
|
| 42 |
+
{"dataset": "imperial_eskd", "comparison": "fewshot", "draw": 3, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.6212115928563943, "comp_l1_by_variant": {"final_cells": 0.6212115928563943, "first_cells": 0.6212115928563943, "final_mrna": 0.7028502433799985, "first_mrna": 0.7028502433799985}, "mean_fraction": {"final_cells": {"T cell": 0.3485, "B cell": 0.0354, "Monocyte": 0.3249, "NK cell": 0.1924, "DC": 0.0526, "Other": 0.0462}, "first_cells": {"T cell": 0.3485, "B cell": 0.0354, "Monocyte": 0.3249, "NK cell": 0.1924, "DC": 0.0526, "Other": 0.0462}, "final_mrna": {"T cell": 0.3052, "B cell": 0.0429, "Monocyte": 0.3638, "NK cell": 0.1459, "DC": 0.0946, "Other": 0.0476}, "first_mrna": {"T cell": 0.3052, "B cell": 0.0429, "Monocyte": 0.3638, "NK cell": 0.1459, "DC": 0.0946, "Other": 0.0476}}, "logfc_r_mean_cp10k": 0.2875792969168268, "partial_r_mean_cp10k": 0.06327553126251119, "logfc_r_mean_cp10k_covered_genes": 0.2875792969168268, "deg_top50_sum": 28, "deg_top50_sum_covered_genes": 28, "per_type_logfc_r": {"T cell": 0.2226, "B cell": 0.092, "Monocyte": 0.6317, "NK cell": 0.204}}, "lora": {"comp_l1": 0.5275199844470723, "coverage": 0.842680564523568, "logfc_r_mean_cp10k": 0.7455905986374256, "deg_top50_sum": 117, "partial_r_mean_cp10k": 0.6007400715066787}, "broadcast": {"logfc_r_mean_cp10k": 0.6538959233815048, "logfc_r_mean_cp10k_covered_genes": 0.6538959233815048, "deg_top50_sum": 76}}
|
| 43 |
+
{"dataset": "imperial_eskd", "comparison": "fewshot", "draw": 4, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.6432264870798222, "comp_l1_by_variant": {"final_cells": 0.6432264870798222, "first_cells": 0.6432264870798222, "final_mrna": 0.7155526059071532, "first_mrna": 0.7155526059071532}, "mean_fraction": {"final_cells": {"T cell": 0.3242, "B cell": 0.0651, "Monocyte": 0.3883, "NK cell": 0.1432, "DC": 0.0479, "Other": 0.0313}, "first_cells": {"T cell": 0.3242, "B cell": 0.0651, "Monocyte": 0.3883, "NK cell": 0.1432, "DC": 0.0479, "Other": 0.0313}, "final_mrna": {"T cell": 0.2912, "B cell": 0.074, "Monocyte": 0.4066, "NK cell": 0.1043, "DC": 0.0829, "Other": 0.0411}, "first_mrna": {"T cell": 0.2912, "B cell": 0.074, "Monocyte": 0.4066, "NK cell": 0.1043, "DC": 0.0829, "Other": 0.0411}}, "logfc_r_mean_cp10k": 0.29331552621576906, "partial_r_mean_cp10k": 0.06931692388975218, "logfc_r_mean_cp10k_covered_genes": 0.29331552621576906, "deg_top50_sum": 37, "deg_top50_sum_covered_genes": 37, "per_type_logfc_r": {"T cell": 0.2391, "B cell": 0.1295, "Monocyte": 0.6495, "NK cell": 0.1552}}, "lora": {"comp_l1": 0.48142813711177446, "coverage": 0.6369376110747358, "logfc_r_mean_cp10k": 0.7492433446082651, "deg_top50_sum": 113, "partial_r_mean_cp10k": 0.575230848198632}, "broadcast": {"logfc_r_mean_cp10k": 0.6565426449733607, "logfc_r_mean_cp10k_covered_genes": 0.6565426449733607, "deg_top50_sum": 76}}
|
| 44 |
+
{"dataset": "imperial_eskd", "comparison": "cv", "draw": null, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.5918271288164071, "comp_l1_by_variant": {"final_cells": 0.5918271288164071, "first_cells": 0.5918271288164071, "final_mrna": 0.6633005066634348, "first_mrna": 0.6633005066634348}, "mean_fraction": {"final_cells": {"T cell": 0.3685, "B cell": 0.07, "Monocyte": 0.324, "NK cell": 0.1331, "DC": 0.0376, "Other": 0.0668}, "first_cells": {"T cell": 0.3685, "B cell": 0.07, "Monocyte": 0.324, "NK cell": 0.1331, "DC": 0.0376, "Other": 0.0668}, "final_mrna": {"T cell": 0.3397, "B cell": 0.0751, "Monocyte": 0.3483, "NK cell": 0.0983, "DC": 0.0674, "Other": 0.0712}, "first_mrna": {"T cell": 0.3397, "B cell": 0.0751, "Monocyte": 0.3483, "NK cell": 0.0983, "DC": 0.0674, "Other": 0.0712}}, "logfc_r_mean_cp10k": 0.22231317619779134, "partial_r_mean_cp10k": 0.03173655156106995, "logfc_r_mean_cp10k_covered_genes": 0.22231317619779134, "deg_top50_sum": 1, "deg_top50_sum_covered_genes": 1, "per_type_logfc_r": {"T cell": 0.1634, "B cell": 0.0189, "Monocyte": 0.4332, "NK cell": 0.2737}}, "lora": {"comp_l1": 0.37638560875745664, "coverage": 0.8500127757431224, "logfc_r_mean_cp10k": 0.8187448084273437, "deg_top50_sum": 143, "partial_r_mean_cp10k": 0.6899314750539278}, "broadcast": {"logfc_r_mean_cp10k": 0.6606053132085035, "logfc_r_mean_cp10k_covered_genes": 0.6606053132085035, "deg_top50_sum": 84}}
|
| 45 |
+
/home/sj_server_1/miniforge3/envs/cxg_corpus/lib/python3.11/site-packages/scipy/stats/_axis_nan_policy.py:635: RuntimeWarning: Precision loss occurred in moment calculation due to catastrophic cancellation. This occurs when the data are nearly identical. Results may be unreliable.
|
| 46 |
+
res = hypotest_fun_out(*samples, axis=axis, **kwds)
|
| 47 |
+
/home/sj_server_1/miniforge3/envs/cxg_corpus/lib/python3.11/site-packages/scipy/stats/_axis_nan_policy.py:635: RuntimeWarning: Precision loss occurred in moment calculation due to catastrophic cancellation. This occurs when the data are nearly identical. Results may be unreliable.
|
| 48 |
+
res = hypotest_fun_out(*samples, axis=axis, **kwds)
|
| 49 |
+
{"dataset": "giroux2022", "comparison": "fewshot", "draw": 0, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.8513318718371892, "comp_l1_by_variant": {"final_cells": 0.8513318718371892, "first_cells": 0.8513318718371892, "final_mrna": 0.5719201437614763, "first_mrna": 0.5719201437614763}, "mean_fraction": {"final_cells": {"T cell": 0.2944, "B cell": 0.054, "Monocyte": 0.1323, "NK cell": 0.0428, "DC": 0.0018, "Other": 0.4747}, "first_cells": {"T cell": 0.2944, "B cell": 0.054, "Monocyte": 0.1323, "NK cell": 0.0428, "DC": 0.0018, "Other": 0.4747}, "final_mrna": {"T cell": 0.3339, "B cell": 0.0781, "Monocyte": 0.2639, "NK cell": 0.056, "DC": 0.007, "Other": 0.2612}, "first_mrna": {"T cell": 0.3339, "B cell": 0.0781, "Monocyte": 0.2639, "NK cell": 0.056, "DC": 0.007, "Other": 0.2612}}, "logfc_r_mean_cp10k": 0.04316512775105488, "partial_r_mean_cp10k": 0.03361235769608455, "logfc_r_mean_cp10k_covered_genes": 0.04316512775105488, "deg_top50_sum": 0, "deg_top50_sum_covered_genes": 0, "per_type_logfc_r": {"T cell": NaN, "B cell": NaN, "Monocyte": 0.0432, "NK cell": NaN}}, "lora": {"comp_l1": 0.27193277679338645, "coverage": 0.96, "logfc_r_mean_cp10k": 0.243392212681724, "deg_top50_sum": 0, "partial_r_mean_cp10k": 0.15035969319388034}, "broadcast": {"logfc_r_mean_cp10k": 0.4709654147613206, "logfc_r_mean_cp10k_covered_genes": 0.4709654147613206, "deg_top50_sum": 0}}
|
| 50 |
+
/home/sj_server_1/miniforge3/envs/cxg_corpus/lib/python3.11/site-packages/scipy/stats/_axis_nan_policy.py:635: RuntimeWarning: Precision loss occurred in moment calculation due to catastrophic cancellation. This occurs when the data are nearly identical. Results may be unreliable.
|
| 51 |
+
res = hypotest_fun_out(*samples, axis=axis, **kwds)
|
| 52 |
+
/home/sj_server_1/miniforge3/envs/cxg_corpus/lib/python3.11/site-packages/scipy/stats/_axis_nan_policy.py:635: RuntimeWarning: Precision loss occurred in moment calculation due to catastrophic cancellation. This occurs when the data are nearly identical. Results may be unreliable.
|
| 53 |
+
res = hypotest_fun_out(*samples, axis=axis, **kwds)
|
| 54 |
+
{"dataset": "giroux2022", "comparison": "fewshot", "draw": 1, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.5781759971117253, "comp_l1_by_variant": {"final_cells": 0.5781759971117253, "first_cells": 0.5781759971117253, "final_mrna": 0.44474798825029893, "first_mrna": 0.44474798825029893}, "mean_fraction": {"final_cells": {"T cell": 0.4133, "B cell": 0.0285, "Monocyte": 0.179, "NK cell": 0.041, "DC": 0.005, "Other": 0.3333}, "first_cells": {"T cell": 0.4133, "B cell": 0.0285, "Monocyte": 0.179, "NK cell": 0.041, "DC": 0.005, "Other": 0.3333}, "final_mrna": {"T cell": 0.4252, "B cell": 0.0381, "Monocyte": 0.3121, "NK cell": 0.0475, "DC": 0.0163, "Other": 0.1608}, "first_mrna": {"T cell": 0.4252, "B cell": 0.0381, "Monocyte": 0.3121, "NK cell": 0.0475, "DC": 0.0163, "Other": 0.1608}}, "logfc_r_mean_cp10k": NaN, "partial_r_mean_cp10k": NaN, "logfc_r_mean_cp10k_covered_genes": NaN, "deg_top50_sum": 0, "deg_top50_sum_covered_genes": 0, "per_type_logfc_r": {"T cell": NaN, "B cell": NaN, "Monocyte": NaN, "NK cell": NaN}}, "lora": {"comp_l1": 0.22320689222847964, "coverage": 0.9526666666666667, "logfc_r_mean_cp10k": 0.35634959703435354, "deg_top50_sum": 1, "partial_r_mean_cp10k": 0.2523250741286013}, "broadcast": {"logfc_r_mean_cp10k": 0.5012527613870964, "logfc_r_mean_cp10k_covered_genes": 0.5012527613870964, "deg_top50_sum": 0}}
|
| 55 |
+
/home/sj_server_1/miniforge3/envs/cxg_corpus/lib/python3.11/site-packages/scipy/stats/_axis_nan_policy.py:635: RuntimeWarning: Precision loss occurred in moment calculation due to catastrophic cancellation. This occurs when the data are nearly identical. Results may be unreliable.
|
| 56 |
+
res = hypotest_fun_out(*samples, axis=axis, **kwds)
|
| 57 |
+
/home/sj_server_1/miniforge3/envs/cxg_corpus/lib/python3.11/site-packages/scipy/stats/_axis_nan_policy.py:635: RuntimeWarning: Precision loss occurred in moment calculation due to catastrophic cancellation. This occurs when the data are nearly identical. Results may be unreliable.
|
| 58 |
+
res = hypotest_fun_out(*samples, axis=axis, **kwds)
|
| 59 |
+
{"dataset": "giroux2022", "comparison": "fewshot", "draw": 2, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.7362234910018671, "comp_l1_by_variant": {"final_cells": 0.7362234910018671, "first_cells": 0.7362234910018671, "final_mrna": 0.5549823156091591, "first_mrna": 0.5549823156091591}, "mean_fraction": {"final_cells": {"T cell": 0.3282, "B cell": 0.0622, "Monocyte": 0.1744, "NK cell": 0.0325, "DC": 0.0089, "Other": 0.3939}, "first_cells": {"T cell": 0.3282, "B cell": 0.0622, "Monocyte": 0.1744, "NK cell": 0.0325, "DC": 0.0089, "Other": 0.3939}, "final_mrna": {"T cell": 0.3593, "B cell": 0.0877, "Monocyte": 0.2967, "NK cell": 0.0432, "DC": 0.0238, "Other": 0.1894}, "first_mrna": {"T cell": 0.3593, "B cell": 0.0877, "Monocyte": 0.2967, "NK cell": 0.0432, "DC": 0.0238, "Other": 0.1894}}, "logfc_r_mean_cp10k": 0.14196498056245352, "partial_r_mean_cp10k": 0.036577277293133814, "logfc_r_mean_cp10k_covered_genes": 0.14196498056245352, "deg_top50_sum": 0, "deg_top50_sum_covered_genes": 0, "per_type_logfc_r": {"T cell": 0.142, "B cell": NaN, "Monocyte": NaN, "NK cell": NaN}}, "lora": {"comp_l1": 0.2907494763185344, "coverage": 0.9533333333333334, "logfc_r_mean_cp10k": 0.230513969445487, "deg_top50_sum": 0, "partial_r_mean_cp10k": 0.15040912208962715}, "broadcast": {"logfc_r_mean_cp10k": 0.39643909018360535, "logfc_r_mean_cp10k_covered_genes": 0.39643909018360535, "deg_top50_sum": 0}}
|
| 60 |
+
/home/sj_server_1/miniforge3/envs/cxg_corpus/lib/python3.11/site-packages/scipy/stats/_axis_nan_policy.py:635: RuntimeWarning: Precision loss occurred in moment calculation due to catastrophic cancellation. This occurs when the data are nearly identical. Results may be unreliable.
|
| 61 |
+
res = hypotest_fun_out(*samples, axis=axis, **kwds)
|
| 62 |
+
/home/sj_server_1/miniforge3/envs/cxg_corpus/lib/python3.11/site-packages/scipy/stats/_axis_nan_policy.py:635: RuntimeWarning: Precision loss occurred in moment calculation due to catastrophic cancellation. This occurs when the data are nearly identical. Results may be unreliable.
|
| 63 |
+
res = hypotest_fun_out(*samples, axis=axis, **kwds)
|
| 64 |
+
{"dataset": "giroux2022", "comparison": "fewshot", "draw": 3, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.6273383088965232, "comp_l1_by_variant": {"final_cells": 0.6273383088965232, "first_cells": 0.6273383088965232, "final_mrna": 0.5349181669787998, "first_mrna": 0.5349181669787998}, "mean_fraction": {"final_cells": {"T cell": 0.4205, "B cell": 0.0449, "Monocyte": 0.1614, "NK cell": 0.0285, "DC": 0.0044, "Other": 0.3403}, "first_cells": {"T cell": 0.4205, "B cell": 0.0449, "Monocyte": 0.1614, "NK cell": 0.0285, "DC": 0.0044, "Other": 0.3403}, "final_mrna": {"T cell": 0.4199, "B cell": 0.0634, "Monocyte": 0.2956, "NK cell": 0.0351, "DC": 0.0151, "Other": 0.1709}, "first_mrna": {"T cell": 0.4199, "B cell": 0.0634, "Monocyte": 0.2956, "NK cell": 0.0351, "DC": 0.0151, "Other": 0.1709}}, "logfc_r_mean_cp10k": 0.10120488249523606, "partial_r_mean_cp10k": 0.015580317281135174, "logfc_r_mean_cp10k_covered_genes": 0.10120488249523606, "deg_top50_sum": 0, "deg_top50_sum_covered_genes": 0, "per_type_logfc_r": {"T cell": 0.1401, "B cell": NaN, "Monocyte": 0.0623, "NK cell": NaN}}, "lora": {"comp_l1": 0.2688359990061283, "coverage": 0.9624242424242424, "logfc_r_mean_cp10k": 0.24186295766157903, "deg_top50_sum": 0, "partial_r_mean_cp10k": 0.14697551642627654}, "broadcast": {"logfc_r_mean_cp10k": 0.4657480400911531, "logfc_r_mean_cp10k_covered_genes": 0.4657480400911531, "deg_top50_sum": 0}}
|
| 65 |
+
/home/sj_server_1/miniforge3/envs/cxg_corpus/lib/python3.11/site-packages/scipy/stats/_axis_nan_policy.py:635: RuntimeWarning: Precision loss occurred in moment calculation due to catastrophic cancellation. This occurs when the data are nearly identical. Results may be unreliable.
|
| 66 |
+
res = hypotest_fun_out(*samples, axis=axis, **kwds)
|
| 67 |
+
/home/sj_server_1/miniforge3/envs/cxg_corpus/lib/python3.11/site-packages/scipy/stats/_axis_nan_policy.py:635: RuntimeWarning: Precision loss occurred in moment calculation due to catastrophic cancellation. This occurs when the data are nearly identical. Results may be unreliable.
|
| 68 |
+
res = hypotest_fun_out(*samples, axis=axis, **kwds)
|
| 69 |
+
{"dataset": "giroux2022", "comparison": "fewshot", "draw": 4, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.5538402679700511, "comp_l1_by_variant": {"final_cells": 0.5538402679700511, "first_cells": 0.5538402679700511, "final_mrna": 0.3915609545174028, "first_mrna": 0.3915609545174028}, "mean_fraction": {"final_cells": {"T cell": 0.4197, "B cell": 0.0291, "Monocyte": 0.1543, "NK cell": 0.0698, "DC": 0.0168, "Other": 0.3103}, "first_cells": {"T cell": 0.4197, "B cell": 0.0291, "Monocyte": 0.1543, "NK cell": 0.0698, "DC": 0.0168, "Other": 0.3103}, "final_mrna": {"T cell": 0.436, "B cell": 0.034, "Monocyte": 0.2491, "NK cell": 0.0749, "DC": 0.0467, "Other": 0.1592}, "first_mrna": {"T cell": 0.436, "B cell": 0.034, "Monocyte": 0.2491, "NK cell": 0.0749, "DC": 0.0467, "Other": 0.1592}}, "logfc_r_mean_cp10k": NaN, "partial_r_mean_cp10k": NaN, "logfc_r_mean_cp10k_covered_genes": NaN, "deg_top50_sum": 0, "deg_top50_sum_covered_genes": 0, "per_type_logfc_r": {"T cell": NaN, "B cell": NaN, "Monocyte": NaN, "NK cell": NaN}}, "lora": {"comp_l1": 0.2274734478129418, "coverage": 0.9766666666666667, "logfc_r_mean_cp10k": 0.41085265458568265, "deg_top50_sum": 0, "partial_r_mean_cp10k": 0.24279747694617004}, "broadcast": {"logfc_r_mean_cp10k": 0.6280870820074911, "logfc_r_mean_cp10k_covered_genes": 0.6280870820074911, "deg_top50_sum": 0}}
|
| 70 |
+
{"dataset": "giroux2022", "comparison": "cv", "draw": null, "gene_mode": "cibersortx", "bayesprism": {"comp_l1": 0.5914100321768722, "comp_l1_by_variant": {"final_cells": 0.5914100321768722, "first_cells": 0.5914100321768722, "final_mrna": 0.4536133148837987, "first_mrna": 0.4536133148837987}, "mean_fraction": {"final_cells": {"T cell": 0.4474, "B cell": 0.024, "Monocyte": 0.1585, "NK cell": 0.0474, "DC": 0.0025, "Other": 0.3202}, "first_cells": {"T cell": 0.4474, "B cell": 0.024, "Monocyte": 0.1585, "NK cell": 0.0474, "DC": 0.0025, "Other": 0.3202}, "final_mrna": {"T cell": 0.4645, "B cell": 0.0298, "Monocyte": 0.2756, "NK cell": 0.0547, "DC": 0.0078, "Other": 0.1677}, "first_mrna": {"T cell": 0.4645, "B cell": 0.0298, "Monocyte": 0.2756, "NK cell": 0.0547, "DC": 0.0078, "Other": 0.1677}}}, "lora": {"comp_l1": 0.2580442754044374, "coverage": 0.9280555555555555}, "broadcast": {}}
|
| 71 |
+
# CIBERSORTx with the same paired donors vs LoRA (paired benchmark; gene mode cibersortx)
|
| 72 |
+
|
| 73 |
+
CIBERSORTx in place of BayesPrism ('BP' columns): official Docker Fractions, single-cell reference (<=400 cells per type from the same paired donors) with S-mode batch correction, then HiRes per-sample cell-type expression on the scoring genes (tools/run_cibersortx.sh). One fraction estimate, so 'final' = 'first'. HiRes needs more samples than cell types: splits with <= 8 test samples (Ji and Giroux CV folds) have composition only.
|
| 74 |
+
|
| 75 |
+
Same donors, observed truth, contrast and scorer as the benchmark (score_ft.py functions; the recomputed bulk broadcast equals the stored one in every split). BayesPrism reference = the split's paired donors' observed cells (states = gate | AIFI L2); mixture = the test samples' real bulk. Composition L1 in cell units (lower is better): 'final' and 'first' BayesPrism theta. logFC r = case-vs-control per-type logFC (mean CP10K) correlation with observed, mean of T/B/Mono/NK; BayesPrism Z zero-filled on genes it lacks ('covered' = only genes it models, broadcast recomputed there). DEG = top-50 overlap summed over types.
|
| 76 |
+
|
| 77 |
+
## Five-donor draws (mean ± SD over 5 draws)
|
| 78 |
+
|
| 79 |
+
| cohort | L1 BP final | L1 BP first | L1 LoRA-5 | r BP | r BP (covered) | r LoRA-5 | r broadcast | partial r BP | partial r LoRA-5 | DEG BP | DEG LoRA-5 | DEG broadcast |
|
| 80 |
+
|---|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|
|
| 81 |
+
| ImMicro (5) | 0.491 ± 0.095 | 0.491 ± 0.095 | 0.176 ± 0.012 | – | – | – | – | – | – | – | – | – |
|
| 82 |
+
| Sepsis (5) | 0.582 ± 0.024 | 0.582 ± 0.024 | 0.466 ± 0.044 | 0.069 ± 0.023 | 0.069 ± 0.023 | 0.350 ± 0.162 | 0.320 ± 0.024 | -0.010 ± 0.026 | 0.305 ± 0.189 | 23.2 ± 2.2 | 16.4 ± 6.7 | 20.0 ± 1.9 |
|
| 83 |
+
| Ji ChILI (5) | 0.827 ± 0.145 | 0.827 ± 0.145 | 0.513 ± 0.115 | 0.158 ± 0.054 | 0.158 ± 0.054 | 0.333 ± 0.077 | 0.234 ± 0.026 | 0.005 ± 0.031 | 0.300 ± 0.076 | 0.0 ± 0.0 | 0.2 ± 0.4 | 1.0 ± 2.2 |
|
| 84 |
+
| Imperial (5) | 0.646 ± 0.026 | 0.646 ± 0.026 | 0.503 ± 0.022 | 0.283 ± 0.011 | 0.283 ± 0.011 | 0.749 ± 0.024 | 0.656 ± 0.009 | 0.061 ± 0.014 | 0.603 ± 0.033 | 30.6 ± 3.9 | 113.6 ± 6.3 | 74.8 ± 7.7 |
|
| 85 |
+
| Giroux (5) | 0.669 ± 0.124 | 0.669 ± 0.124 | 0.256 ± 0.030 | 0.095 ± 0.050 | 0.095 ± 0.050 | 0.297 ± 0.082 | 0.492 ± 0.085 | 0.029 ± 0.011 | 0.189 ± 0.054 | 0.0 ± 0.0 | 0.2 ± 0.4 | 0.0 ± 0.0 |
|
| 86 |
+
|
| 87 |
+
Draws where LoRA-5 beats CIBERSORTx: composition 24/25 (vs CIBERSORTx's better theta), logFC r 17/17, partial r 18/18, top-50 DEG 8/20 (ties 8).
|
| 88 |
+
|
| 89 |
+
## 5-fold CV, pooled (CIBERSORTx reference = each fold's LoRA training + inner-validation pairs)
|
| 90 |
+
|
| 91 |
+
| cohort | L1 BP final | L1 BP first | L1 LoRA | L1 scSemi | r BP | r BP (covered) | r LoRA | r scSemi | r broadcast | partial r BP | partial r LoRA | DEG BP | DEG LoRA | DEG scSemi | DEG broadcast |
|
| 92 |
+
|---|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|
|
| 93 |
+
| ImMicro | 0.507 | 0.507 | 0.162 | 0.217 | – | – | – | – | – | – | – | – | – | – | – |
|
| 94 |
+
| Sepsis | 0.607 | 0.607 | 0.362 | 0.502 | -0.045 | -0.045 | 0.527 | 0.435 | 0.316 | -0.081 | 0.469 | 0 | 23 | 6 | 21 |
|
| 95 |
+
| Ji ChILI | 0.914 | 0.914 | 0.504 | 0.575 | – | – | – | 0.273 | – | – | – | – | – | 0 | – |
|
| 96 |
+
| Imperial | 0.592 | 0.592 | 0.376 | – | 0.222 | 0.222 | 0.819 | – | 0.661 | 0.032 | 0.690 | 1 | 143 | – | 84 |
|
| 97 |
+
| Giroux | 0.591 | 0.591 | 0.258 | 0.243 | – | – | – | 0.531 | – | – | – | – | – | 6 | – |
|
| 98 |
+
{"dataset": "ye2026_sepsis", "comparison": "cv", "draw": null, "gene_mode": "full", "bayesprism": {"comp_l1": 0.6029270442284831, "comp_l1_by_variant": {"final_cells": 0.6029270442284831, "first_cells": 0.4750523423944086, "final_mrna": 0.649204630549072, "first_mrna": 0.45424984326877027}, "mean_fraction": {"final_cells": {"T cell": 0.3123, "B cell": 0.1357, "Monocyte": 0.2537, "NK cell": 0.2517, "DC": 0.0064, "Other": 0.0401}, "first_cells": {"T cell": 0.6081, "B cell": 0.0653, "Monocyte": 0.2181, "NK cell": 0.0623, "DC": 0.0041, "Other": 0.0421}, "final_mrna": {"T cell": 0.2639, "B cell": 0.1304, "Monocyte": 0.3495, "NK cell": 0.2163, "DC": 0.015, "Other": 0.0248}, "first_mrna": {"T cell": 0.5244, "B cell": 0.0659, "Monocyte": 0.3155, "NK cell": 0.0569, "DC": 0.0102, "Other": 0.0271}}, "logfc_r_mean_cp10k": 0.2554670801971232, "partial_r_mean_cp10k": 0.09471562043764327, "logfc_r_mean_cp10k_covered_genes": 0.2533386702661269, "deg_top50_sum": 37, "deg_top50_sum_covered_genes": 38, "per_type_logfc_r": {"T cell": 0.1232, "B cell": 0.1825, "Monocyte": 0.5804, "NK cell": 0.1358}}, "lora": {"comp_l1": 0.3618246783833512, "coverage": 0.829349593495935, "logfc_r_mean_cp10k": 0.5267612711229187, "deg_top50_sum": 23, "partial_r_mean_cp10k": 0.46874787884206276, "deg_source": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/scores/ye2026_sepsis/pooled_deg/summary.json"}, "broadcast": {"logfc_r_mean_cp10k": 0.31594581626202867, "logfc_r_mean_cp10k_covered_genes": 0.2531577574908379, "deg_top50_sum": 21}}
|
| 99 |
+
# BayesPrism with the same paired donors vs LoRA (paired benchmark; gene mode full)
|
| 100 |
+
|
| 101 |
+
BayesPrism genes: its recommended set (canonical protein-coding minus chrX/Y/M and RPS/RPL/MRPS/MRPL, measured by the sc assay; 15-19k genes).
|
| 102 |
+
|
| 103 |
+
Same donors, observed truth, contrast and scorer as the benchmark (score_ft.py functions; the recomputed bulk broadcast equals the stored one in every split). BayesPrism reference = the split's paired donors' observed cells (states = gate | AIFI L2); mixture = the test samples' real bulk. Composition L1 in cell units (lower is better): 'final' and 'first' BayesPrism theta. logFC r = case-vs-control per-type logFC (mean CP10K) correlation with observed, mean of T/B/Mono/NK; BayesPrism Z zero-filled on genes it lacks ('covered' = only genes it models, broadcast recomputed there). DEG = top-50 overlap summed over types.
|
| 104 |
+
|
| 105 |
+
## Five-donor draws (mean ± SD over 5 draws)
|
| 106 |
+
|
| 107 |
+
| cohort | L1 BP final | L1 BP first | L1 LoRA-5 | r BP | r BP (covered) | r LoRA-5 | r broadcast | partial r BP | partial r LoRA-5 | DEG BP | DEG LoRA-5 | DEG broadcast |
|
| 108 |
+
|---|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|
|
| 109 |
+
| ImMicro (5) | 0.594 ± 0.088 | 0.550 ± 0.164 | 0.176 ± 0.012 | – | – | – | – | – | – | – | – | – |
|
| 110 |
+
| Sepsis (5) | 0.643 ± 0.118 | 0.543 ± 0.218 | 0.466 ± 0.044 | 0.241 ± 0.035 | 0.258 ± 0.036 | 0.350 ± 0.162 | 0.320 ± 0.024 | 0.095 ± 0.035 | 0.305 ± 0.189 | 27.0 ± 2.0 | 16.4 ± 6.7 | 20.0 ± 1.9 |
|
| 111 |
+
| Ji ChILI (5) | 0.943 ± 0.117 | 0.952 ± 0.161 | 0.513 ± 0.115 | 0.193 ± 0.013 | 0.212 ± 0.013 | 0.333 ± 0.077 | 0.234 ± 0.026 | 0.032 ± 0.020 | 0.300 ± 0.076 | 2.0 ± 2.4 | 0.2 ± 0.4 | 1.0 ± 2.2 |
|
| 112 |
+
| Imperial (5) | 0.962 ± 0.086 | 0.845 ± 0.063 | 0.503 ± 0.022 | 0.598 ± 0.033 | 0.614 ± 0.034 | 0.749 ± 0.024 | 0.656 ± 0.009 | 0.187 ± 0.039 | 0.603 ± 0.033 | 72.4 ± 10.1 | 113.6 ± 6.3 | 74.8 ± 7.7 |
|
| 113 |
+
| Giroux (5) | 1.807 ± 0.047 | 1.455 ± 0.361 | 0.256 ± 0.030 | 0.371 ± 0.060 | 0.374 ± 0.060 | 0.297 ± 0.082 | 0.492 ± 0.085 | 0.117 ± 0.060 | 0.189 ± 0.054 | 0.2 ± 0.4 | 0.2 ± 0.4 | 0.0 ± 0.0 |
|
| 114 |
+
|
| 115 |
+
Draws where LoRA-5 beats BayesPrism: composition 22/25 (vs BayesPrism's better theta), logFC r 15/20, partial r 18/20, top-50 DEG 6/20 (ties 6).
|
| 116 |
+
|
| 117 |
+
## 5-fold CV, pooled (BayesPrism reference = each fold's LoRA training + inner-validation pairs)
|
| 118 |
+
|
| 119 |
+
| cohort | L1 BP final | L1 BP first | L1 LoRA | L1 scSemi | r BP | r BP (covered) | r LoRA | r scSemi | r broadcast | partial r BP | partial r LoRA | DEG BP | DEG LoRA | DEG scSemi | DEG broadcast |
|
| 120 |
+
|---|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|---:|
|
| 121 |
+
| ImMicro | 0.575 | 0.909 | 0.162 | 0.217 | – | – | – | – | – | – | – | – | – | – | – |
|
| 122 |
+
| Sepsis | 0.603 | 0.475 | 0.362 | 0.502 | 0.255 | 0.253 | 0.527 | 0.435 | 0.316 | 0.095 | 0.469 | 37 | 23 | 6 | 21 |
|
| 123 |
+
| Ji ChILI | 0.883 | 1.055 | 0.504 | 0.575 | 0.186 | 0.204 | 0.479 | 0.273 | 0.240 | 0.020 | 0.452 | 4 | 15 | 0 | 4 |
|
| 124 |
+
| Imperial | 0.887 | 1.022 | 0.376 | – | 0.568 | 0.584 | 0.819 | – | 0.661 | 0.112 | 0.690 | 91 | 143 | – | 84 |
|
| 125 |
+
| Giroux | 1.803 | 0.878 | 0.258 | – | 0.406 | 0.408 | 0.461 | – | 0.575 | 0.022 | 0.323 | 3 | 5 | – | 5 |
|
| 126 |
+
2026-10-08T10:57:59 finalize_cx done; failed=0
|