File size: 2,717 Bytes
ff5bd45 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 | ---
pretty_name: Final Directional Co-Mutation Dependency Atlas
language:
- en
tags:
- cancer-genomics
- depmap
- crispr
- mutation
- drug-response
- reproducible-research
---
# Final Directional Co-Mutation Dependency Atlas
**Standalone fresh computation**: every statistical result was recomputed from official raw DepMap 24Q4 Public v1, PRISM 19Q4 v4, Reactome and optional cited sources. No outputs, C1 shortlist, or checkpoints from historical projects were loaded.
The source code is under `src/final_comutation/`; experiment-section figures and tables are under `results/Paper_Experiment_Section/`. Computation status and limitations are under `results/Audit/`.
## Findings for this particular run (dynamic)
- C1 conservative discoveries: 83
- Selected across-context results: 546
- Drug inference: {"causal_mechanism_established": false, "interpretation": "Conditional drug response support differs from A-by-B nonadditive interaction or clinical therapeutic effect", "n_AB_vs_Bonly_BH": 0, "n_AB_vs_Bonly_tested": 50, "n_PRISM_BH_concordant": 8, "n_PRISM_BY_concordant": 5, "n_PRISM_conditional_tested": 49, "n_annotated_drug_hypotheses": 62, "n_external_independently_validated": 0, "n_fresh_C1": 83, "n_strict_AxB_BH": 0, "n_strict_interactions_tested": 31, "raw_data_only_no_prior_results": true, "study": "FRESH raw-source drug-response followup"}
## Data provenance and scope
Input providers: DepMap 24Q4 v1 (versioned figshare 27993248 v1), PRISM Repurposing 19Q4 v4 (figshare 9393293 v4), pinned Reactome 98, optionally Enrichr/MSigDB oncogenic signatures, cBioPortal selected TCGA cancer studies, and optional identified GDSC2. The raw input matrices and per-cell data are **not redistributed** here; the original providers should be cited and their terms reviewed.
### Interpretation safeguards
- CRISPR and drug-response contrasts are observational; co-mutation states do not establish causal order or genetic synergy.
- A=mutated and B=additionally mutated versus A-only is the primary estimand. A×B and B-only contrasts are auxiliary.
- Multiplicity-adjusted associations are computational, not independent cohort or wet-lab validation.
- `No reported qualifying mutation` is **not** verified wild type.
- TCGA prevalence is not patient treatment response.
- Failed/underpowered and unavailable validation evidence is retained explicitly.
- Authors must inspect all exported evidence and methods before manuscript submission.
## Reproducibility
Run the provided multi-cell Kaggle notebook or `python -m final_comutation.pipeline`. Source code and source manifests are preserved; see `results/Paper_Experiment_Section/Methods_and_Evidence/Experiment_Section_Claim_Evidence_Methods.md`.
|