OpenAI, PerturbAI and Scripps Research develop multi-agent biology research framework

OpenAI, PerturbAI and Scripps Research develop multi-agent biology research framework

The framework uses nearly 2,000 AI agents to analyze biological experiment data, search literature and generate hypotheses for later testing. PerturbAI says it analyzed its whole-brain genome atlas and integrated research findings in 32 minutes. In a blind comparison of reports, experts preferred the multi-agent system's explanations 75% of the time. That preference does not establish the accuracy of its conclusions, and the research leads still need experimental validation.

OpenAI, PerturbAI and Scripps Research use AI agents to analyze a mouse brain gene experiment

The three organizations had 1,946 GPT-Rosalind-5.5 agents analyze a whole-brain mouse CRISPR atlas in parallel in 32 minutes. The source says they matched or refined 48 of 55 published findings supported by the expression data. The system also proposed a hypothesis about cholesterol transport and neurodegeneration that remains to be tested.

OpenAI, PerturbAI, and Scripps Research use AI agents to seek research leads in brain cell data

OpenAI, PerturbAI, and Scripps Research used distributed AI agents to seek promising research leads in data from millions of brain cells. The material says thousands of agents were used, but does not say whether the leads were experimentally validated.