GNN explanation methods reveal disease signatures in biological networks

Researchers have evaluated four popular explanation methods for graph neural networks (GNNs) to understand their effectiveness in identifying disease-associated structures within biological networks. Using synthetic data and breast cancer RNA sequencing data, the study found that different methods excel at uncovering distinct types of biological signals, such as single-node drivers or distributed pathways. By combining consensus scores from multiple explainers and incorporating topological information, the researchers improved the prioritization of key cancer genes and the recovery of biologically relevant signaling pathways. AI

IMPACT Improves biological interpretability of GNNs, potentially leading to more accurate disease diagnosis and drug discovery.

RANK_REASON The cluster contains an academic paper detailing a new methodology for interpreting graph neural network outputs in a biological context. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• BRCA1
• ERBB2
• ESR1
• GNNExplainer
• Graph neural networks
• Integrated Gradients
• Layer-wise Relevance Propagation
• Saliency Attribution
• TCGA BRCA data
• TP53

• paper
• safety

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →