Exploratory action

ECxit

Exiting the EC Classification for Better Enzyme Annotation by Deep Learning
Exiting the EC Classification for Better Enzyme Annotation by Deep Learning

Deep language models, such as those behind ChatGPT, have revolutionized natural language processing. By treating protein sequences as a language, the ECxit project aims to transfer these advances to the field of biology. Its goal is to develop a novel method and a redesigned classification of enzymes, enabling their identification and the precise prediction of their functions from amino acid sequences, ultimately improving genome annotation.

Inria teams involved

DYLISS

Contacts

François Coste

Scientific leader