MotifDiv

MotifDiv is a dataset of conformational-specificity design challenges from the PDB. It consists of 200 pairs of single-domain proteins with high structural homology, except for a 10-residue motif with high structural divergence. It is suited for the evaluation of methods to design protein sequences with high conformational specificity.

The full dataset generation protocol is described in our paper here: https://www.biorxiv.org/content/10.1101/2025.04.23.650138v1