Is Biomni's code execution safe?

Biomni executes LLM-generated code with system privileges. For security, it is highly recommended to run the agent in isolated environments like Docker or virtual machines.

What specific biological tasks can it perform?

It excels at CRISPR screening design, single-cell RNA-seq analysis, ADMET prediction for drugs, GWAS interpretation, and clinical phenotype-genotype mapping.

Biomni is an autonomous biomedical AI agent framework from Stanford SNAP lab that executes complex research tasks across genomics, drug discovery, and clinical analysis.

How does Biomni handle biomedical data?

It integrates an 11GB data lake containing gene databases (Ensembl, NCBI), protein structures, clinical datasets (ClinVar, OMIM), and literature indices which are automatically downloaded on first use.

Do I need to write code to use Biomni?

No, Biomni can autonomously generate and execute the necessary Python code to perform data analysis based on your natural language research queries.

Biomni Biomedical Research Agent

Name: Biomni Biomedical Research Agent
Author: Zehong-Wang

byZehong-Wang

0•

Data Science & ML

Executes autonomous multi-step biomedical research tasks including genomics analysis, drug discovery, and CRISPR design.

Biomni is an open-source framework from Stanford's SNAP lab designed to function as an autonomous AI agent for complex biomedical research. It bridges the gap between high-level biological reasoning and technical execution by automatically decomposing tasks, retrieving data from an integrated 11GB biomedical database, and generating or executing code for complex data analysis. Whether you are performing single-cell RNA sequencing, predicting drug ADMET properties, or interpreting GWAS results, Biomni provides a unified interface to streamline end-to-end research workflows across genomics, proteomics, and clinical domains within your development environment.

Key Features

01Support for specialized tasks like CRISPR screening, scRNA-seq, and GWAS interpretation

02Automated Python code generation and execution for data analysis pipelines

03Autonomous multi-step task decomposition and planning for complex research queries

040 GitHub stars

05Integrated access to 11GB of curated biomedical databases, literature, and ontologies

06Cross-domain reasoning across genomics, proteomics, and drug discovery

Use Cases

01Performing end-to-end single-cell RNA-seq analysis from quality control to cell annotation

02Designing and prioritizing genes for genome-wide CRISPR knockout screens

03Predicting ADMET properties and identifying targets for drug discovery candidates

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add zehong-wang/kosmos biomni

For use in Claude.ai and ChatGPT

Download Skill