Differentiable Gene Set Enrichment Analysis for Pathway-Level Supervision in Transcriptomic Learning

Read full text Summarise with VIRAT
Authors: Shuaiyu Li, Yang Ruan, Xinyue Yang, Zhang Wen, Hiroto Saigo
Cited by: 0
References: 33

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Abstract In transcriptomics-driven drug discovery, upstream predictors of chemical-induced transcriptional profiles (CTPs) are typically trained with gene-wise objectives, whereas downstream interpretation relies on pathway-level, rank-based statistics such as Gene Set Enrichment Analysis (GSEA). This objective mismatch destabilizes pathway conclusions under prediction errors: small ranking perturbations can flip enrichment direction or distort pathway ordering. To bridge this gap, we present differentiable GSEA (dGSEA), a training-compatible surrogate that maps predicted gene-level scores to pathway enrichment with well-behaved gradients. Technically, dGSEA replaces discrete ranking operations with temperature-controlled soft sorting, smooth prefix accumulation, and differentiable extremum aggregation. Critically, to preserve the statistical semantics of classical GSEA, we introduce sign-specific robust permutation normalization (dNES) with optional κ -calibration. For computational efficiency, a scalable Nyström–window approximation (nyswin) reduces the quadratic bottleneck to near-linear complexity, enabling genome-scale evaluation. Empirically, across synthetic benchmarks and LINCS L1000 signatures, dGSEA matches classical GSEA accuracy with improved numerical stability. When incorporated as an auxiliary objective for SMILES-to-transcriptome prediction, dGSEA improves pathway-level agreement (macro correlation 0.257 → 0.306; sign accuracy 0.620 → 0.641) without compromising gene-level performance, providing a practical mechanism for pathway-aware optimization in transcriptomic prediction pipelines.

Fields of Study & Subjects

Bioinformatics and Genomic Networks Computational Drug Discovery Methods Gene Regulatory Network Analysis

Publication details

Journal / source
bioRxiv (Cold Spring Harbor Laboratory)
Publication date
20 March 2026
Language
EN
Document type
Preprint