A group of researchers funded by the Arc Institute unveiled on Wednesday an innovative artificial intelligence (AI) model, Evo 2, which incorporates the genomes of all living species. This ambitious project aims to contribute to the development of new medical treatments and advance research in biology.
An unprecedented genomic database
Evo 2 is based on a colossal database containing more than 128,000 complete genomes. This immense amount of information is hosted on Amazon Web Services (AWS) cloud servers, guaranteeing optimal accessibility and computing power.
The development of this AI model is part of a strategic collaboration with Nvidia, a leader in the semiconductor industry. To develop this advanced interface, no fewer than 2,000 H100 chips, Nvidia’s flagship product, were used.
Accelerating Research on Human Diseases
This partnership with Evo 2 offers an innovative tool that can accelerate the understanding of complex human diseases. Silvana Konermann, director of the Arc Institute, emphasizes that this model will make it possible to identify the genetic variants involved in various pathologies. The goal is to be able to create new molecules that precisely target these areas to improve treatments.
Tests conducted on a gene associated with breast cancer illustrate the potential of this tool. Evo 2 has demonstrated a 90% reliable prediction capability, distinguishing between benign and pathogenic genetic mutations.
- AI with Multiple Applications Beyond its role in medical research, Evo 2 opens new perspectives in other scientific fields. Its potential includes:
- Agriculture: improving plant resilience to extreme climatic conditions and enhancing their nutritional value. Ecology
- : contributing to the creation of enzymes capable of breaking down plastic, thus helping to combat pollution.
Pharmacology
: identifying new medicinal targets by predicting potential molecular interactions.
A large-scale collaborative project
The Evo 2 project brings together researchers from several renowned Californian universities, including Stanford, UC Berkeley, and UCSF. The Arc Institute offers them the opportunity to work at the institute while maintaining their ties with academia, thus fostering a dynamic and collaborative research environment.
