dyMEAN and MEAN
These are successive versions of the same method, where dyMEAN extends MEAN from conditional 3D graph translation of antibody backbones to end-to-end full-atom design, making the earlier version largely superseded.
About dyMEAN
THUNLP-MT/dyMEAN
This repo contains the codes for our paper "End-to-End Full-Atom Antibody Design"
This project provides an end-to-end system for designing full-atom antibody structures, focusing on crucial binding regions and overall protein stability. It takes existing antibody structural data (PDB files) and, through various modules, outputs optimized antibody designs with improved binding affinity and predicted complex structures. Researchers and scientists in drug discovery, immunology, and protein engineering fields can use this to accelerate the development of new therapeutic antibodies.
About MEAN
THUNLP-MT/MEAN
This repo contains the codes for our paper Conditional Antibody Design as 3D Equivariant Graph Translation.
This project helps biological engineers and immunologists design antibodies by predicting optimal protein sequences for specific binding sites. You provide the 3D structure of an existing antibody-antigen complex, and it generates new amino acid sequences for the antibody's Complementarity-Determining Region (CDR-H3) that enhance binding or improve other properties. This tool is ideal for researchers in antibody engineering, drug discovery, or protein design.
Scores updated daily from GitHub, PyPI, and npm data. How scores work