Progressive Multi-Agent Reasoning for Biological Perturbation Prediction

2026-02-07

AI Summary

This work addresses the challenge of predicting target gene regulatory responses in bulk-cell populations under complex chemical perturbations, where existing methods struggle to model the causal entanglement among high-dimensional perturbations. The authors propose PBio-Agent, a multi-agent framework that introduces the first benchmark for bulk-perturbation response prediction, termed LINCSQA, and incorporates a causal structure sharing assumption. The framework features difficulty-aware task sequencing and iterative knowledge refinement, with specialized agents enhanced by biological knowledge graphs to enable collaborative reasoning. A synthesis agent integrates predictions while a verification agent ensures logical consistency. Evaluated on LINCSQA and PerturbQA, PBio-Agent significantly outperforms current approaches, substantially improving both performance and interpretability of smaller models in complex biological perturbation prediction tasks.

Technology Category

• Multiagent Systems: Modeling other Agents
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Knowledge Representation and Reasoning: Action, Change, and Causality

Application Category

• 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
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Predicting gene regulation responses to biological perturbations requires reasoning about underlying biological causalities. While large language models (LLMs) show promise for such tasks, they are often overwhelmed by the entangled nature of high-dimensional perturbation results. Moreover, recent works have primarily focused on genetic perturbations in single-cell experiments, leaving bulk-cell chemical perturbations, which is central to drug discovery, largely unexplored. Motivated by this, we present LINCSQA, a novel benchmark for predicting target gene regulation under complex chemical perturbations in bulk-cell environments. We further propose PBio-Agent, a multi-agent framework that integrates difficulty-aware task sequencing with iterative knowledge refinement. Our key insight is that genes affected by the same perturbation share causal structure, allowing confidently predicted genes to contextualize more challenging cases. The framework employs specialized agents enriched with biological knowledge graphs, while a synthesis agent integrates outputs and specialized judges ensure logical coherence. PBio-Agent outperforms existing baselines on both LINCSQA and PerturbQA, enabling even smaller models to predict and explain complex biological processes without additional training.

Problem

• biological perturbation prediction
• gene regulation
• chemical perturbations
• bulk-cell
• causal reasoning

Innovation

• multi-agent reasoning
• biological perturbation prediction
• knowledge graph integration
• iterative knowledge refinement
• bulk-cell chemical perturbation