Washington - ALsharqiya, March 20: Google announced on Tuesday the development of a new set of open-source artificial intelligence models called TxGemma, which aims to accelerate the drug discovery process. These advanced models are capable of understanding plain text and handling various therapeutic combinations such as chemicals, molecules, and proteins.
Google said it will launch these models later this month through the Health AI Developer Foundations program. Karen DiSalvo, who leads Google's technology in medicine team, said that the process of developing drugs from concept to approved use is a long and expensive process, noting that the company is working with the broader research community to find new ways to speed up this process.
DiSalvo added that researchers can use TxGemma to ask questions that help them predict the characteristics of potential new treatments, such as their safety and effectiveness. Despite these innovations, Google has not revealed whether the models' licenses will allow for commercial use, customization, or modification. Many companies, including Google's Isomorphic Labs, have promised that AI will revolutionize drug discovery by accelerating early stages of research and development. Despite some successes in this area, companies using AI in drug discovery have recently experienced some high-profile failures in clinical trials. Pioneering systems such as Google DeepMind's AlphaFold 3 have made progress in this area, albeit with significant variations in accuracy.
