Domain-Adversarial Neural Network and Explainable AI for Reducing Tissue-of-Origin Signal in Pan-cancer Mortality Classification
Preprint
Abstract
Tissue-of-origin signals dominate pan-cancer gene expression, often obscuring molecular features linked to patient survival. This hampers the discovery of generalizable biomarkers, as models tend to overfit tissue-specific patterns rather than capture survival-relevant signals. To address this, we propose a Domain-Adversarial Neural Network (DANN) trained on TCGA RNA-seq data to learn representations less biased by tissue and more focused on survival. Identifying tissue-independent genetic profiles is key to revealing core cancer programs. We assess the DANN using: (1) Standard SHAP, based on the original input space and DANN's mortality classifier; (2) A layer-aware strategy applied to hidden activations, including an unsupervised manifold from raw activations and a supervised manifold from mortality-specific SHAP values. Standard SHAP remains confounded by tissue signals due to biases inherent in its computation. The raw activation manifold was dominated by high-magnitude activations, which masked subtle tissue and mortality-related signals. In contrast, the layer-aware SHAP manifold offers improved low-dimensional representations of both tissue and mortality signals, independent of activation strength, enabling subpopulation stratification and pan-cancer identification of survival-associated genes.Keywords
All Related Versions
This publication has 49 references indexed in Scilit:
- Deep learning for unsupervised domain adaptation in medical imaging: Recent advancements and future perspectivesComputers in Biology and Medicine, 2023
- Comprehensive analysis of RNA binding motif protein 3 (RBM3) in non‐small cell lung cancerCancer Medicine, 2020
- Prospective oncotarget for gynecological cancer: Opioid growth factor (OGF) - opioid growth factor receptor (OGFr) axisInternational Immunopharmacology, 2019
- PRKDC is a prognostic marker for poor survival in gastric cancer patients and regulates DNA damage responsePathology - Research and Practice, 2019
- RPS7 promotes cell migration through targeting epithelial-mesenchymal transition in prostate cancerUrologic Oncology: Seminars and Original Investigations, 2019
- Overexpression of KLHL23 protein from read‐through transcription of PHOSPHO2‐KLHL23 in gastric cancer increases cell proliferationFEBS Open Bio, 2016
- CXCL9 and CXCL10 predict survival and are regulated by cyclooxygenase inhibition in advanced serous ovarian cancerBritish Journal of Cancer, 2016
- High RNA-binding motif protein 3 expression is an independent prognostic marker in operated prostate cancer and tightly linked to ERG activation and PTEN deletionsEuropean Journal of Cancer, 2014
- The Cancer Genome Atlas Pan-Cancer analysis projectNature Genetics, 2013
- Global estimates of cancer prevalence for 27 sites in the adult population in 2008International Journal of Cancer, 2012