Contributing annotations of predicted intrinsic protein disorder to the RCSB Protein Data Bank

Abstract

The Protein Data Bank (PDB) holds information about three‐dimensional (3D) structures of biological macromolecules. It’s a global resource used by industry, researchers, students and educators for various ends. To serve users with diverse needs, the RCSB PDB (accessible at rcsb.org) integrates additional information associated with each structure, among which are properties derived from sequence data. Predictions of highly dynamic disordered regions are available for PDB entries, calculated using JRONN, a Java port of RONN algorithm. The IUPred2A method for calculations of disordered protein regions is evaluated to replace the previously used JRONN method. Previously published benchmark studies show that IUPred2A is more accurate than JRONN. By pre‐computing disordered regions and serving this information with each protein sequence in the PDB, users gain additional reference information and adds context to solved protein structures without requiring users to derive property information on their own. Furthermore, the integration of disorder calculations can be combined with other reference data on the PDB. For example, many entries in the PDB are highly structured globular molecules, but the addition of predicted regions of disorder can reveal short unstructured regions that can be important for protein‐protein interactions. Calculations of intrinsically disordered regions in protein sequences contributes valuable reference information to the PDB and can serve diverse research needs.Support or Funding InformationNational Science Foundation (DBI‐1832184), the US Department of Energy (DE‐SC0019749), and the National Cancer Institute, National Institute of Allergy and Infectious Diseases, and National Institute of General Medical Sciences of the National Institute of Health under grant R01GM133198.