Biomni Biomedical AI Agent

关于

Biomni is an open-source framework from Stanford's SNAP lab designed to function as an autonomous AI agent for multi-step biomedical research. It bridges the gap between high-level reasoning and technical execution by autonomously decomposing tasks, generating analysis code, and querying over 11GB of integrated biomedical data. This skill is ideal for researchers and bioinformaticians needing to streamline specialized workflows like CRISPR screen design, single-cell RNA-seq analysis, or drug ADMET prediction within their Claude environment.

主要功能

• Multi-step biological reasoning and autonomous task planning
• Integrated access to 11GB of biomedical databases including NCBI and UniProt
• Cross-domain support for genomics, proteomics, and drug discovery
• Built-in evaluation framework for benchmarking biomedical research outputs
• 81 GitHub stars
• Dynamic analysis pipeline creation via code generation and execution

使用场景

• Designing genome-wide CRISPR screens and prioritizing gene targets
• Predicting ADMET properties and identifying targets for drug discovery
• Performing end-to-end single-cell RNA-seq clustering and annotation