Comprehensive Quantum Chemical Interaction Analysis of Whole Protein–Ligand Structures in BindingDB Using the Fragment Molecular Orbital Method
7 affiliations
- 1Graduate School of Pharmaceutical Sciences
- 2University of Osaka
- 3Computational Chemistry Business Department, Engineering Technology Business Unit
- 4JSOL Corporation
- 5Department of Applied Chemistry, Graduate School of Engineering
- 6Kyushu University
- 7Graduate School of Information Science and Electrical Engineering
Abstract
Understanding protein–ligand interactions based on their electronic states is important for elucidating molecular recognition and supporting structure-based drug discovery. In this study, we performed fragment molecular orbital (FMO) calculations, a quantum chemical approach, for a comprehensive set of experimentally determined protein–ligand complexes derived from BindingDB. Ligand–amino acid residue interactions were quantitatively characterized using inter-fragment interaction energies (IFIEs) and pair interaction energy decomposition analysis (PIEDA), and were compared with molecular mechanics (MM) interaction energies. The comparison revealed residue- and interaction-dependent differences between the FMO and MM descriptions, with particularly distinct behavior for interactions involving aromatic residues. Comprehensive IFIE/PIEDA analysis further revealed characteristic energetic contributions associated with amino acid residue type and ligand charge, including dispersion contributions involving aromatic residues and charge-transfer contributions associated with hydrogen-bonding interactions. A case study of MAPK14 demonstrated that the resulting dataset can identify residue regions commonly involved in ligand recognition and distinguish residue-level interaction patterns among individual ligands. These results provide a systematic quantum chemical perspective on diverse protein–ligand interactions and establish a quantum chemical resource for complementing structural, experimental, and conventional MM-based analyses.
Related
- Interaction of nitric oxide with the (1 0 0) surface of cobalt spinel nanocubes – A comprehensive DFT, atomistic thermodynamic, IR and TPD account
- Chemical Structures Cleaning Center: An Open Web Service for Automated Chemical Databases Curation
- Two extrapolation axes for many-body systems: interaction without an interaction module, chemical mixtures as the main example
- Orthonym: verified IUPAC names for chemical structures in the wild
- Interaction between S-phase precipitates and dislocation sub-structures in AA2024 (Al-Cu-Mg)
BibTeX
@misc{miyakawa2026comprehensive,
title = {{Comprehensive Quantum Chemical Interaction Analysis of Whole Protein–Ligand Structures in BindingDB Using the Fragment Molecular Orbital Method}},
author = {Shuhei Miyakawa and Naoko Konami and Ruri Mihata and Rikuto Osaki and Ayaka Masuda and Tomoka Taguchi and Toma Miyagishi and Koji Okuwaki and Koichiro Kato and Yoshiharu Mori and Yu-Shi Tian and Daisuke Takaya and Kaori Fukuzawa},
howpublished = {ChemRxiv},
year = {2026},
month = sep,
doi = {10.26434/chemrxiv.15009601/v1},
url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15009601/v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Comprehensive Quantum Chemical Interaction Analysis of Whole Protein–Ligand Structures in BindingDB Using the Fragment Molecular Orbital Method
AU - Shuhei Miyakawa
AU - Naoko Konami
AU - Ruri Mihata
AU - Rikuto Osaki
AU - Ayaka Masuda
AU - Tomoka Taguchi
AU - Toma Miyagishi
AU - Koji Okuwaki
AU - Koichiro Kato
AU - Yoshiharu Mori
AU - Yu-Shi Tian
AU - Daisuke Takaya
AU - Kaori Fukuzawa
PY - 2026
DA - 2026/09/29
PB - ChemRxiv
DO - 10.26434/chemrxiv.15009601/v1
UR - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15009601/v1
AB - Understanding protein–ligand interactions based on their electronic states is important for elucidating molecular recognition and supporting structure-based drug discovery. In this study, we performed fragment molecular orbital (FMO) calculations, a quantum chemical approach, for a comprehensive set of experimentally determined protein–ligand complexes derived from BindingDB. Ligand–amino acid residue interactions were quantitatively characterized using inter-fragment interaction energies (IFIEs) and pair interaction energy decomposition analysis (PIEDA), and were compared with molecular mechanics (MM) interaction energies. The comparison revealed residue- and interaction-dependent differences between the FMO and MM descriptions, with particularly distinct behavior for interactions involving aromatic residues. Comprehensive IFIE/PIEDA analysis further revealed characteristic energetic contributions associated with amino acid residue type and ligand charge, including dispersion contributions involving aromatic residues and charge-transfer contributions associated with hydrogen-bonding interactions. A case study of MAPK14 demonstrated that the resulting dataset can identify residue regions commonly involved in ligand recognition and distinguish residue-level interaction patterns among individual ligands. These results provide a systematic quantum chemical perspective on diverse protein–ligand interactions and establish a quantum chemical resource for complementing structural, experimental, and conventional MM-based analyses.
ER -