Lemon: a framework for rapidly mining structural information from the Protein Data Bank

Abstract

MotivationThe protein data bank (PDB) currently holds over 140,000 biomolecular structures and continues to release new structures on a weekly basis. The PDB is an essential resource to the structural bioinformatics community to develop software that mine, use, categorize, and analyze such data. New computational biology methods are evaluated using custom benchmarking sets derived as subsets of 3D experimentally determined structures and structural features from the PDB. Currently, such benchmarking features are manually curated with custom scripts in a non-standardized manner that results in slow distribution and updates with new experimental structures. Finally, there is a scarcity of standardized tools to rapidly query 3D descriptors of the entire PDB. ApproachOur solution is the Lemon framework, a C++11 library with Python bindings, which provides a consistent workflow methodology for selecting biomolecular interactions based on user criterion and computing desired 3D structural features. This framework can parse and characterize the entire PDB in less than ten minutes on modern, multithreaded hardware. The speed in parsing is obtained by using the recently developed MacroMolecule Transmission Format (MMTF) to reduce the computational cost of reading text-based PDB files. The use of C++ lambda functions and Python binds provide extensive flexibility for analysis and categorization of the PDB by allowing the user to write custom functions to suite their objective. We think Lemon will become a one-stop-shop to quickly mine the entire PDB to generate desired structural biology features. The Lemon software is available as a C++ header library along with example functions at https://github.com/chopralab/lemon.