Google Cloud has recently announced a groundbreaking partnership with Ginkgo Bioworks, introducing two new releases that will revolutionize the world of drug discovery and artificial intelligence in biology.
The first release includes the launch of a protein large language model (LLM) that will provide organizations and companies with the tools to expedite their drug discovery process using Ginkgo’s proprietary insights and data. The second release is a model API created by Ginkgo Bioworks to grant machine learning engineers and scientists access to cutting-edge biology AI models.
In an exclusive interview with Forbes, Chris Sakalosky, Vice President of Strategic Industries at Google Cloud, and Ankit Gupta, General Manager of Ginkgo AI, shared their insights on the significance of these new developments. Gupta emphasized the shift in biology from a physical discipline to a computational one, highlighting the importance of leveraging data for success. Sakalosky discussed the parallels between DNA and language, suggesting that AI has the potential to understand and work with DNA in a similar manner as it does with spoken language and computer code.
The protein LLM, built on Google Cloud’s Vertex AI platform and trained on Ginkgo’s data models, holds great promise for researchers and enterprises in accelerating data analysis for drug discovery. Additionally, the model API offers access to sophisticated models trained on protein and DNA data, such as the ginkgo-AA-0-650m model trained on over 2 billion proprietary Ginkgo protein sequences.
The convergence of biology, drug discovery, artificial intelligence, and advanced engineering presents vast opportunities for growth and innovation. Companies like Meta with the ESM Metagenomic Atlas and Alphabet’s Isomorphic Labs with AlphaFold 3 are making significant strides in leveraging AI for protein structure prediction and analysis. Studies in bioinformatics are already recognizing the potential of LLMs in solving complex problems in medicine and molecular biology.
While there is still much work to be done in this field, the advancements in protein modeling and AI integration offer a promising glimpse into the future of biological sciences. The collaboration between Google Cloud and Ginkgo Bioworks signifies a significant step towards harnessing the power of data and technology for groundbreaking discoveries in drug development and beyond.
