GALAX: Graph-Augmented Language Model for Explainable Reinforcement-Guided Subgraph Reasoning in Precision Medicine

2025-09-25

AI Summary

Current approaches to multi-omics pathway and target identification in precision medicine suffer from insufficient topological awareness, inadequate quantitative reasoning, underutilization of node-level semantic information, and poor generalizability of large language models (LLMs). To address these bottlenecks, we propose GALAX: a novel framework that introduces the Graph-enhanced Process Reward Model (GPRM) for unsupervised, stepwise subgraph generation—enabling interpretable graph-text joint reasoning. We further construct Target-QA, a benchmark supporting long-context, text-numerical-graph trimodal reasoning. GALAX synergistically integrates pretrained graph neural networks (GNNs) and LLMs via reinforcement learning to jointly model multi-omics features, protein–protein interaction (PPI) topology, and literature-scale biological knowledge. In cancer cell line experiments, GALAX achieves significant improvements in biological plausibility, cross-dataset generalizability, and mechanistic interpretability of identified therapeutic targets.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Natural Language Processing: Language Grounding & Multi-modal NLP
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

In precision medicine, quantitative multi-omic features, topological context, and textual biological knowledge play vital roles in identifying disease-critical signaling pathways and targets. Existing pipelines capture only part of these-numerical omics ignore topological context, text-centric LLMs lack quantitative grounded reasoning, and graph-only models underuse node semantics and the generalization of LLMs-limiting mechanistic interpretability. Although Process Reward Models (PRMs) aim to guide reasoning in LLMs, they remain limited by unreliable intermediate evaluation, and vulnerability to reward hacking with computational cost. These gaps motivate integrating quantitative multi-omic signals, topological structure with node annotations, and literature-scale text via LLMs, using subgraph reasoning as the principle bridge linking numeric evidence, topological knowledge and language context. Therefore, we propose GALAX (Graph Augmented LAnguage model with eXplainability), an innovative framework that integrates pretrained Graph Neural Networks (GNNs) into Large Language Models (LLMs) via reinforcement guided by a Graph Process Reward Model (GPRM), which generates disease-relevant subgraphs in a step-wise manner initiated by an LLM and iteratively evaluated by a pretrained GNN, enabling process-level supervision without explicit intermediate reasoning annotations. As an application, we also introduced Target-QA, a benchmark combining CRISPR-identified targets, multi-omic profiles, and biomedical graph knowledge across diverse cancer cell lines, which enables GNN pretraining for supervising step-wise graph construction and supports long-context reasoning over text-numeric graphs (TNGs), providing a scalable and biologically grounded framework for explainable, reinforcement-guided subgraph reasoning toward reliable and interpretable target and pathway discovery in precision medicine.

Problem

• Integrating quantitative multi-omic signals with topological structure and textual knowledge
• Addressing limitations of existing models in mechanistic interpretability and grounded reasoning
• Developing explainable subgraph reasoning for reliable target and pathway discovery

Innovation

• Integrates pretrained GNNs into LLMs via reinforcement learning
• Uses a Graph Process Reward Model for step-wise subgraph generation
• Enables process-level supervision without intermediate reasoning annotations