SLInterpreter: An Exploratory and Iterative Human-AI Collaborative System for GNN-based Synthetic Lethal Prediction

TL;DR

SLInterpreter is a human-AI collaborative system that improves GNN-based synthetic lethal prediction by integrating expert knowledge, interpretive path exploration, and iterative refinement to enhance biological understanding and trust.

Contribution

It introduces an iterative framework combining knowledge graph refinement and multi-granularity interpretation to align AI predictions with domain expertise in cancer gene research.

Findings

Enhanced interpretability of GNN predictions

Improved alignment with biological domain knowledge

Facilitated discovery of new synthetic lethal relationships

Abstract

Synthetic Lethal (SL) relationships, though rare among the vast array of gene combinations, hold substantial promise for targeted cancer therapy. Despite advancements in AI model accuracy, there is still a significant need among domain experts for interpretive paths and mechanism explorations that align better with domain-specific knowledge, particularly due to the high costs of experimentation. To address this gap, we propose an iterative Human-AI collaborative framework with two key components: 1) Human-Engaged Knowledge Graph Refinement based on Metapath Strategies, which leverages insights from interpretive paths and domain expertise to refine the knowledge graph through metapath strategies with appropriate granularity. 2) Cross-Granularity SL Interpretation Enhancement and Mechanism Analysis, which aids experts in organizing and comparing predictions and interpretive paths across different granularities, uncovering new SL relationships, enhancing result interpretation, and elucidating potential mechanisms inferred by Graph Neural Network (GNN) models. These components cyclically optimize model predictions and mechanism explorations, enhancing expert involvement and intervention to build trust. Facilitated by SLInterpreter, this framework ensures that newly generated interpretive paths increasingly align with domain knowledge and adhere more closely to real-world biological principles through iterative Human-AI collaboration. We evaluate the framework's efficacy through a case study and expert interviews.