SeekBrain: An Autonomous Multi-Agent System for Accelerating Neuroscience Discovery

2026-07-31

AI Summary

This work addresses the challenges of integrating multiscale, multimodal data and fragmented analytical workflows in neuroscience by proposing the first autonomous multi-agent analysis framework. The framework integrates analysis protocols automatically extracted from code-paper pairs with a hierarchical task planning mechanism, enabling dynamic generation of scalable, on-demand analytical pipelines through domain knowledge–guided cross-modal data fusion. Evaluated on the BrainArena benchmark, it significantly outperforms existing methods and successfully uncovers the distributed neural representations underlying zebrafish behavior as well as a shared axis of decoding strength across brain regions during mouse decision-making tasks, thereby advancing the automation and systematicity of discovery in neuroscience.

Abstract

Modern neuroscience relies on integrating multi-scale, multimodal datasets to uncover the neural principles underlying intelligence. However, analytical challenges posed by highly heterogeneous data and fragmented workflows increasingly constrain discoveries. Here we introduce SeekBrain, an autonomous multi-agent framework designed to accelerate neuroscience discovery through domain-grounded hierarchical planning and cross-modal data analysis. SeekBrain dynamically constructs a repertoire of analysis recipes extracted from code-paper pairs. By coupling this codified expertise with agentic planning and execution engines, the framework scalably generates hypotheses and analytical pipelines on demand. Systematic evaluation on the expert-annotated BrainArena benchmark demonstrates that SeekBrain substantially outperforms state-of-the-art agent baselines across various analysis tasks. Crucially, when deployed in real-world research, SeekBrain integrated behavioral, neural, and anatomical data to reveal structured, distributed neural representations of larval zebrafish behavior and a shared axis of regional decoding strength across the brain in a mouse decision-making task. These results establish SeekBrain as a scalable and practical tool for accelerating data-driven discoveries in neuroscience.

Problem

• neuroscience
• heterogeneous data
• multimodal data
• fragmented workflows
• data integration

Innovation

• autonomous multi-agent system
• cross-modal data analysis
• hierarchical planning
• neuroscience discovery
• analysis recipe generation