Home Search Benchmark AlphaFold Tutorial Reference Contact Version 1.0


A web-based tool for real-time protein surface comparison and analysis. The server integrates a repertoire of methods to assist in high throughput screening and visualization of protein surface comparisons. 3D-Zernike Descriptors (3DZD) are utilized for the efficient comparison between a protein surface of a single chain, a single domain, or a single complex against all protein structures in the latest Protein Data Bank (PDB). Conveniently, the web interface also renders animated protein rotations, displays CATH protein structure classification codes, and structure alignment calculations using the Combinatorial Extension (CE) algorithm, and identifies geometric (i.e., pocket) regions on a given structure.

Now AlphaFold models are included for a search!

Statistics of latest release (Last updated on Oct 20, 2021):

AlphaFold Models 508,787
All PDB Entries 893,546
  Chain Entries 547,639
  Complex Entries 96,744
  Domain Entries 249,163
All (CACN) Entries 893,512
  Chain (CACN) Entries 547,605
  Complex (CACN) Entries 96,744
  Domain (CACN) Entries 249,163


Latest features:

  • AlphaFold models included: AlphaFold models downloaded from the AlphaFold Protein Structure Database at EBI are now included for structure search. A search can be performed with a deep neural network that was trained for retrieving proteins of the same fold.
  • Surface representation: Two options of surface representation are provided. The first one is the all-atom-surface representation, and the second is the backbone-atom-surface representation, which includes alpha C (CA), C, and N atoms in the main chain.

  • Database: 3DZD of protein chains, domains, complexes and a combination of three above are repertoried in the website.

  • Batch mode: A batch mode is available to run multiple proteins for uses who would like to benchmark their methods against our approach.

  • Maintenance: The databases are weekly updated in pace with the latest PDB.

Copyright © 2021 KIHARA Bioinformatics LABORATORY, PURDUE University