Recent Developments of FMODB in 2025: Enhancing FMO Data Accessibility through Visualization Tools
Kikuko KAMISAKA *1, Chiduru WATANABE1, Shu KOYAMA2, Kazumi TSUDA2, Gert-Jan BEKKER3, Genji KURISU3, Teruki HONMA1
1Center for Integrative Medical Sciences, RIKEN
2Science & Technology Systems, Inc.
3Protein Data Bank Japan, Institute for Protein Research, University of Osaka
The Fragment Molecular Orbital (FMO) method is a quantum chemical approach that enables high-precision component-wise decomposition of interaction energies in biomolecular systems. It has been increasingly applied in drug discovery and structural biology. Since 2017, under the leadership of the FMO Drug Design Consortium (FMODD), we have continuously developed and maintained FMODB (https://drugdesign.riken.jp/FMODB/), a database that systematically manages and shares FMO calculation results.Quantitative understanding of intermolecular interactions is essential for molecular design in drug discovery and functional analysis in structural biology. The interfragment interaction energies (IFIEs) and their decomposition (PIEDA) provided by FMO are effective for evaluating intermolecular binding modes and residue-level contributions, thereby aiding binding site analysis. FMODB serves as a platform that delivers these high-precision data in a consistent and accessible format.
This presentation introduces recent updates aimed at promoting the use of FMODB and expanding its applications. As of July 2025, FMODB contains approximately 44,500 FMO entries including about 7,100 added in 2025. We are also working to support AlphaFold-predicted models and non-protein materials such as amorphous solid dispersions.In addition, PDBj’s high-performance 3D molecular viewer Molmil now supports PIEDA visualization and the FMODB web interface has been updated to allow 3D mapping of FMO results. This combination of features enables users to visually identify within FMODB, which residues contribute to interactions through specific physicochemical factors (e.g., electrostatics, dispersion, charge transfer), facilitating binding site optimization and functional hotspot analysis.
Furthermore, we implemented data linkage from FMODB to the FMO calculation support GUI library FMOe, enabling direct retrieval and loading of CPF files for streamlined visualization and analysis. FMOe runs on MOE (Molecular Operating Environment) and supports fragment division manipulation, input preparation, IFIE/PIEDA visualization, charge distribution display and automatic fragmentation of nucleic acids.Through these enhancements, FMODB aims to promote further application of the FMO method by supporting structure-based research with FMO data.
Acknowledgment
This work was conducted as part of the activities of the FMO Drug Design Consortium (https://fmodd.jp/top-en/) and was partially supported by the Research Support Project for Life Science and Drug Discovery (Basis for Supporting Innovative Drug Discovery and Life Science Research [BINDS]) from AMED under Grant Number JP25ama121030 and JSPS KAKENHI Grant Number 23K18192. FMO calculations were performed using the supercomputer Fugaku (project IDs: hp250154 and ra250009).
[1] Takaya, D. et al., J. Chem. Inf. Model. 2021, 61, 2, 777–794.
[2] Bekker, G.-J. et al., J. Cheminform. 2016, 8 (1), 42.
[3] Moriwaki, H. et al., J. Chem. Inf. Model. 2024, 64, 18.