Reproducible analysis package for stress-testing SLC7A11 temporal and cellular associations in hepatocellular carcinoma

Le résumé fourni par la source

Code, compact analysis inputs, locked outputs and figure-generation scripts supporting the manuscript: “Stress-testing the temporal and cellular reproducibility of SLC7A11 associations in hepatocellular carcinoma: a five-cohort and patient-level single-cell reappraisal.” Version 5 reconstructs five outcome-traceable hepatocellular carcinoma survival cohorts comprising 993 patients and 308 deaths. It compares constant, continuous log-time and prespecified two-year piecewise Cox models, retains five-fold cross-validated inverse-probability-of-censoring-weighted Brier diagnostics, and performs patient-level single-cell analyses in GSE149614 and GSE125449. The package includes executable Python code, compact public-data inputs, frozen analysis specifications, cohort- and patient-level outputs, figure-generation scripts, software requirements and a SHA-256 file manifest. GSE36376 is retained only for a documented eligibility audit because reproducible patient-level survival fields could not be recovered from its official public record. The GSE238264 spatial analysis is exploratory. Its large raw Visium archive is not redistributed; the package provides a SHA-256-verified downloader for optional reproduction. Version 5 supersedes version 4 (https://doi.org/10.5281/zenodo.21452878). The analyses use public, de-identified datasets and do not establish causal mechanism, clinical prediction utility or therapeutic efficacy.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.