Machine Learning Models for In Silico Antibody Evolution

Researchers at Stanford have invented a method using directed evolution for predicting antibody improved antibody variants.

Evolution searches across an immense space of possible sequences for rare mutations that improve fitness. In nature, this search is based on random mutation and recombination, but using this same approach for directed evolution of proteins in the laboratory imposes a considerable experimental burden. Indeed, current approaches are limited as high-throughput evolutionary screens, that rely on random guessing or brute-force search, often devote substantial effort to interrogating weakly active or nonfunctional proteins. There is therefore a need for more efficient and effective methods of generating protein variants using directed evolution.

Stage of Development
Research -

in vitro

Stage of Research
The inventors have developed methods for predicting improved antibody variants. Such variants can be used (or selected) to improve properties of known therapeutic antibodies in methods described herein. The present disclosure is also based, in part, on antibody variants designed using the provided methods that are able to specifically bind to viral antigens.

Technology Reference Nos.
Biohub ref. no. CZB-245S-PC
Stanford ref. no. S22-130