Hugging face launches AI model to interact with dna sequences

Hugging Face, the AI platform, has launched Carbon-X, an open-source model enabling users to converse about DNA. Developed with researchers from Zhongguancun Academy in China and the University of Naples Federico II in Italy, the model is trained on genomic sequences and natural language, allowing users to ask questions about DNA changes and receive conversational responses. Lewis Tunstall, a senior research scientist at Hugging Face, highlighted the tool's potential to aid in drug and genetic therapy design, as well as help patients interpret their genomes. Carbon-X integrates DNA and natural language in a conversational interface, akin to ChatGPT for DNA. The model was trained on 300 billion tokens combining DNA sequences, English text, and biological descriptions. It was not trained on viruses to avoid potential misuse. Users can query DNA sequences up to 50,000 base pairs, but cannot upload full human genomes. Tunstall emphasized the model's local execution to ensure data privacy, with no data sent to Hugging Face or other companies. The team warns that the model has not been validated with wet lab experiments, so its accuracy remains unproven.