Artificial intelligence (AI) could play an important role in speeding up laborious methods of finding out what genes do and how they interact, according to a paper published in Nature Methods.
Five different large language models were tested, with GPT-4 found to be the most successful.
It had a 73 percent accuracy rate in scanning a commonly used genomics database to identify the common functions of curated gene sets.
Read more at UC San Diego Today.
