Processed source data and analysis code for evidence-bounded SGTA transcriptomic and proteomic analysis in lung adenocarcinoma

This research compendium contains analysis scripts, processed statistical outputs, figure-level source data, figure previews, provenance records, and audit files supporting a multi-omic study of SGTA in lung adenocarcinoma. Version 1.2.2 retains the complete-pair RNA–protein sensitivity analyses, patient-subsampling network re-discovery, same-tree cluster controls, exact-size discovery-matched anchor-module nulls, public PDC000489 retrieval script, proteome-wide paired-effect context, leave-one-protein-out module-membership sensitivity, collinearity diagnostics, and exploratory tumour-purity and protein-composition sensitivities from v1.2.1. It additionally rebuilds Figure 1 as a result-led cohort evidence map and evidence ledger and adds two plot-ready Figure 1 source tables. No analysis, estimate, eligibility rule, or interpretation changes. The analyses establish reproducible SGTA protein elevation but limit network inference to descriptive edge concordance conditional on frozen discovery membership; they do not establish stable membership or SGTA-specific network organisation. No new primary patient or experimental data were generated. Raw TCGA, GEO, CPTAC/PDC, and DepMap matrices are not redistributed. Original code in code/ is licensed under MIT; study-generated processed outputs, figure source data, figure previews, provenance, and documentation are licensed under CC BY 4.0.Administrative correction (28 July 2026): Outdated authorship metadata in legacy manuscript-building materials were corrected. This administrative correction does not affect the data, analyses, figures, results, or scientific conclusions.