AUTHOR=Eliasof Moshe , Boesen  Tue , Haber  Eldad , Keasar  Chen , Treister  Eran 

TITLE=Mimetic Neural Networks: A Unified Framework for Protein Design and Folding

JOURNAL=Frontiers in Bioinformatics

VOLUME=Volume 2 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/bioinformatics/articles/10.3389/fbinf.2022.715006

DOI=10.3389/fbinf.2022.715006

ISSN=2673-7647

ABSTRACT=Recent advancements in machine learning techniques for protein structure prediction motivate3
better results in its inverse problem – protein design. In this work we introduce a new graph4
mimetic neural network, MimNet, and show that it is possible to build a reversible architecture5
that solves the structure and design problems in tandem, allowing to improve protein backbone6
design when the structure is better estimated. We use the ProteinNet data set and show that the7
state of the art results in protein design can be met and even improved, given recent architectures8
for protein folding.