Archive 4 checkpoints (1-4)

paper-releases/20260928/diffusion/data/source_provenance/giroux2022/pairing/PAIRING_SUMMARY.json

@@ -0,0 +1,171 @@
+{
+  "bulk_gsms": 71,
+  "bulk_subjects_geo": 29,
+  "bulk_count_table_columns": 49,
+  "bulk_count_table_subjects": 22,
+  "bulk_columns_matched_to_GSM": 43,
+  "bulk_columns_without_GSM": [
+    "77DA77",
+    "22CBD6",
+    "4F2EAD",
+    "F8FF1A",
+    "56E219",
+    "3F05F3-14"
+  ],
+  "bulk_GSMs_without_count_column": 28,
+  "bulk_GSMs_without_count_column_by_group": {
+    "exposed": 20,
+    "healthy": 5,
+    "mild": 2,
+    "severe": 1
+  },
+  "sc_gsms": 41,
+  "sc_subjects": 18,
+  "geo_level_subject_day_matches": 39,
+  "geo_level_subject_day_matches_by_group": {
+    "mild": 15,
+    "severe": 13,
+    "exposed": 6,
+    "healthy": 5
+  },
+  "geo_level_subjects": 18,
+  "pairs_total": 26,
+  "pairs_subjects": 10,
+  "pairs_by_group": {
+    "mild": 13,
+    "severe": 13
+  },
+  "pairs_by_day": {
+    "0": 10,
+    "14": 9,
+    "7": 7
+  },
+  "sc_by_pair_status": {
+    "paired": 26,
+    "bulk_GSM_has_no_column_in_author_count_table": 13,
+    "no_bulk_GSM_for_this_subject_day": 2
+  },
+  "sc_unpaired": [
+    {
+      "sc_gsm": "GSM6249237",
+      "sample_id": "450905_d7",
+      "group": "mild",
+      "pair_status": "bulk_GSM_has_no_column_in_author_count_table"
+    },
+    {
+      "sc_gsm": "GSM6249239",
+      "sample_id": "7768E4_d2",
+      "group": "exposed",
+      "pair_status": "no_bulk_GSM_for_this_subject_day"
+    },
+    {
+      "sc_gsm": "GSM6249240",
+      "sample_id": "7768E4_d14",
+      "group": "exposed",
+      "pair_status": "bulk_GSM_has_no_column_in_author_count_table"
+    },
+    {
+      "sc_gsm": "GSM6249241",
+      "sample_id": "7768E4_d28",
+      "group": "exposed",
+      "pair_status": "no_bulk_GSM_for_this_subject_day"
+    },
+    {
+      "sc_gsm": "GSM6249243",
+      "sample_id": "0B943B_d3",
+      "group": "mild",
+      "pair_status": "bulk_GSM_has_no_column_in_author_count_table"
+    },
+    {
+      "sc_gsm": "GSM6249262",
+      "sample_id": "61BBAD_d0",
+      "group": "exposed",
+      "pair_status": "bulk_GSM_has_no_column_in_author_count_table"
+    },
+    {
+      "sc_gsm": "GSM6249263",
+      "sample_id": "61BBAD_d7",
+      "group": "exposed",
+      "pair_status": "bulk_GSM_has_no_column_in_author_count_table"
+    },
+    {
+      "sc_gsm": "GSM6249264",
+      "sample_id": "61BBAD_d14",
+      "group": "exposed",
+      "pair_status": "bulk_GSM_has_no_column_in_author_count_table"
+    },
+    {
+      "sc_gsm": "GSM6249267",
+      "sample_id": "80A16A_d0",
+      "group": "exposed",
+      "pair_status": "bulk_GSM_has_no_column_in_author_count_table"
+    },
+    {
+      "sc_gsm": "GSM6249268",
+      "sample_id": "80A16A_d14",
+      "group": "exposed",
+      "pair_status": "bulk_GSM_has_no_column_in_author_count_table"
+    },
+    {
+      "sc_gsm": "GSM6249272",
+      "sample_id": "E0FA1E_dNA",
+      "group": "healthy",
+      "pair_status": "bulk_GSM_has_no_column_in_author_count_table"
+    },
+    {
+      "sc_gsm": "GSM6249273",
+      "sample_id": "3F3322_dNA",
+      "group": "healthy",
+      "pair_status": "bulk_GSM_has_no_column_in_author_count_table"
+    },
+    {
+      "sc_gsm": "GSM6249274",
+      "sample_id": "94D44E_dNA",
+      "group": "healthy",
+      "pair_status": "bulk_GSM_has_no_column_in_author_count_table"
+    },
+    {
+      "sc_gsm": "GSM6249275",
+      "sample_id": "52BA23_dNA",
+      "group": "healthy",
+      "pair_status": "bulk_GSM_has_no_column_in_author_count_table"
+    },
+    {
+      "sc_gsm": "GSM6249276",
+      "sample_id": "655A91_dNA",
+      "group": "healthy",
+      "pair_status": "bulk_GSM_has_no_column_in_author_count_table"
+    }
+  ],
+  "author_cohort_vs_geo_exposure": {
+    "0BF51C": {
+      "author_cohort": "Moderate",
+      "geo_exposure": "Severe"
+    },
+    "0E1F8E": {
+      "author_cohort": "Moderate",
+      "geo_exposure": "Severe"
+    },
+    "180E1A": {
+      "author_cohort": "Moderate",
+      "geo_exposure": "Severe"
+    },
+    "3F05F3": {
+      "author_cohort": "Moderate",
+      "geo_exposure": "Severe"
+    },
+    "40067F": {
+      "author_cohort": "Moderate",
+      "geo_exposure": "Severe"
+    },
+    "82CCF5": {
+      "author_cohort": "Moderate",
+      "geo_exposure": "Severe"
+    },
+    "B85D75": {
+      "author_cohort": "Moderate",
+      "geo_exposure": "Severe"
+    }
+  },
+  "rule": "A pair is one sc GEX GSM and one bulk GSM with the same GEO subject ID and the same GEO sampling day (title '<subject>, <day>, <group>, <serostatus>' = characteristics subject/time); both GSMs must agree on GEO age, sex, exposure group and serostatus; and the bulk GSM must have a column in the author featureCounts table (column '<subject>-<day>' for infected/exposed subjects, '<subject>' for healthy). No expression, age-sex or row-order matching is used to create a pair."
+}

paper-releases/20260928/diffusion/data/source_provenance/immicro/offset_sample_size_20260914/PROTOCOL.json

@@ -0,0 +1,28 @@
+{
+  "version": "offset_sample_size_v1",
+  "status": "specified_before_feature_results",
+  "question": "How many independent donors are needed to estimate a constant real-bulk to pseudobulk condition offset in healthy ImMicro?",
+  "cohort": "107 strict ImMicro donors with at least 1000 review-pass cells, excluding HS33",
+  "recipe_k": 1000,
+  "pseudobulk_target": "All explicitly indexed captures combined per donor, QC-pass raw sum rescaled by 1000/total_QC_cells; expected-aggregate target, not an actual 1000-cell subsample",
+  "bulk_preprocessing": "Native STAR gene counts primary; SC-training-depth matched counts sensitivity; no length normalization primary",
+  "sensitivity_modes": ["bulk_depthmatched_to_native_expected_pseudo", "both_modalities_depthmatched"],
+  "conditioner": "Existing frozen E2 full-axis cache; exact full20097 transform followed by panel selection; gene and scalar features clipped as original; mean offset added without further clipping for condition-space evaluation only",
+  "sample_sizes": [2, 4, 6, 8, 12, 16, 24, 32, 48, 64, 80, 107],
+  "bootstrap_repeats": 1000,
+  "heldout_resamples": 200,
+  "seed": 20260914,
+  "primary_numerical_endpoint": "95th percentile of gene-offset RMS error, in original training-standardized units, under iid donor bootstrap with replacement against the empirical107 population mean",
+  "operational_rms_tolerance": 0.1,
+  "sensitivity_rms_tolerances": [0.05, 0.2],
+  "minimum_rule": "Smallest tested n meeting threshold and continuing to meet it at every larger tested n; null if none",
+  "numerical_scope": "Common valid gene coordinates for each panel; six scalar channels assessed separately by RMS and max absolute error",
+  "bootstrap_designs": ["same_donors_both_modalities", "independent_donor_samples_per_modality"],
+  "heldout_design": "Seven leave-one-pool-out folds; remaining donors sampled uniformly without replacement in nested prefixes; same queries and107-donor gallery for every n; no test donor enters offset fit",
+  "heldout_primary_metric": "Mean heldout donor squared error on common valid condition genes",
+  "heldout_secondary_metric": "Correct pseudobulk retrieval by Pearson correlation on common valid genes and six scalar features;107 fixed candidates",
+  "performance_stability_rule": "Mean squared error at most1.10 times and retrieval at most0.05 absolute below the all-other-pools reference in at least90 percent of conditional calibration resamples; report aggregate and each heldout pool",
+  "uncertainty": "Bootstrap and resample intervals describe conditional sampling sensitivity, not independent-cohort confidence intervals; seven pool results reported separately",
+  "pool_effect": "Exploratory unbalanced one-way random-effects method-of-moments variance decomposition on paired difference vectors; descriptive only, seven pools",
+  "restrictions": ["No GPU or generator inference", "No claim of universal minimum across protocols or disease cohorts", "No finite-population correction in iid bootstrap stability curve", "Raw acquisition and prior artifacts preserved", "No sample-specific SC metadata input in bulk K/depth conventions", "No multiple-testing significance claim; threshold curves and effect sizes are descriptive"]
+}