PDB-IHM: Methods for Programmatic Deposition of Integrative Structures to the Protein Data Bank

Authors:
Preprint
Publisher website
Google Scholar
Abstract
Integrative structure determination combines varied experimental information with computational modeling. We describe a programmatic workflow connecting the output of these calculations directly to deposition into the Protein Data Bank (PDB), a Worldwide Protein Data Bank Core Archive that now accommodates integrative or hybrid methods structures (PDB-IHM). The python-ihm package generates deposition-ready mmCIF files describing structure models, experimental restraints, modeling protocols, and their provenance. Subsequently, programmatic interfaces support pre-deposition validation, structure deposition, and processing status tracking. We illustrate the workflow using biostructure models derived from three-dimensional electron microscopy (3DEM) and crosslinking mass spectrometry (crosslinking-MS) information. We also demonstrate the workflow scalability through submission of a collection of crosslinking-MS structure models. Making structure validation and deposition machine-accessible in turn allows incorporation into high-throughput and AI-agent-driven structure-determination workflows. The workflow provides a reproducible and scalable pathway from integrative modeling to archival deposition.

This publication has 24 references indexed in Scilit: