BioChatter: AI-Powered Text Analysis for the Life Sciences

Large language models (LLMs) are revolutionizing life science research, but their use often requires specialized expertise. BioChatter, an open-source Python framework, simplifies access to these technologies. The solution, presented in the journal Nature Biotechnology, enables customized AI applications in biomedicine, plant sciences, and the bioeconomy.

BioChatter facilitates the analysis of large text corpora, extracts relevant information from publications, and integrates with existing bioinformatics tools. Initial tests, including those in plant sciences, have shown promising results. As part of the BioChatter Consortium, researchers from IBG-4 contributed to the application of BioChatter in plant sciences.

Future developments include integration with knowledge graphs and expansion through BioGather to process additional types of life science data.

The publication:

Lobentanzer, S., Feng, S., Bruderer, N. et al. A platform for the biomedical application of large language models. Nat Biotechnol 43, 166–169 (2025). https://doi.org/10.1038/s41587-024-02534-3