ssGSEA3.0: Feature-Level Bootstrap Resampling for Robust Assessment of Statistical Significance

ssGSEA3.0 improves ssGSEA/PTM-SEA statistical significance calibration across weighting schemes by using feature-level bootstrap resampling, boosting detection power versus ssGSEA2.0-fixed.

• Why it matters: Existing ssGSEA2.0 approaches either inflate false positives at nonzero weight or restore calibration while losing detection power, leaving a tradeoff for per-sample pathway activity testing.
• What they did: The method replaces feature-label permutation with feature-level bootstrap resampling and evaluates performance on published lung adenocarcinoma and phosphoproteomic perturbation datasets.
• The result: ssGSEA3.0 detects more signals than ssGSEA2.0-fixed, with extra gains for directional PTM-SEA scoring, while adding about a ~1.5- to 2-fold computational cost.

AI summary of the abstract · may contain errors · not medical advice · read the paper.