This research demonstrates a machine learning (ML)-aided pipeline for designing and screening proteins with emergent functions, using the bacterial MinDE system as a model. The study uses an MSA-VAE to generate MinE variants, a divide-and-conquer computational screening approach based on predicted sub-functions, and in vitro screening using synthetic cell mimics. A high-scoring variant fully substituted the wild-type MinE gene in E. coli, highlighting the potential of this integrated approach for engineering cellular functions.
Publisher
Nature Communications
Published On
Mar 05, 2024
Authors
Shunshi Kohyama, Béla P. Frohn, Leon Babl, Petra Schwille
Tags
machine learning
protein engineering
MinDE system
synthetic biology
E. coli
Related Publications
Explore these studies to deepen your understanding of the subject.
