Protein Structure from Scientific Research

Learning the 3Di alphabet. (1) 3Di states describe tertiary interaction between a residue i and its nearest neighbor j . Nearest neighbors have the closest virtual center distance (yellow). Virtual center positions (Supplementary Fig. 1) were optimized for maximum search sensitivity. (2) To describe the interaction geometry of residues i and j , we extract seven angles, the Euclidean C α distance and two sequence distance features from the six C α coordinates of the two backbone fragments (blue and red). (3) These 10 features are used to define 20 3Di states by training a VQ-VAE 28 modified to learn states that are maximally evolutionary conserved. For structure searches, the encoder predicts the best-matching 3Di state for each residue.

DOI