Developments and Activities in FMODB up to 2025: A Release Note

Abstract

The fragment molecular orbital (FMO) method enables quantum chemical evaluation of intra- and intermolecular interactions in biomacromolecules at the fragment level. To facilitate data reuse and advance research in structural biology and drug discovery, we have developed the FMO database (FMODB, URL: https://drugdesign.riken.jp/FMODB/), which currently contains 77,277 entries. In this paper, we summarize the key features added to FMODB since 2021, including advanced search capabilities, enhanced interfragment interaction energy (IFIE) analysis tools, a batch IFIE analysis module, a Web API, and cross-links to PDBj. These updates markedly enhance usability and interoperability, thereby enabling more effective application of FMO data.

1. Introduction

The fragment molecular orbital (FMO) method is a quantum chemical approach for biomacromolecules that enables the evaluation of interactions at the fragment level, such as amino acid residues, nucleic acids, and ligands [1–4]. Based on this methodology, our group has developed the FMO database (FMODB), a web-based resource that systematically archives and provides FMO calculation results [5,6]. The development of FMODB started in 2017, followed by its public release in 2019 [7]. In 2021, we reported the basic design of the database and its initial data contents in a dedicated publication [5]. Building on this foundation, the present work focuses on major new features implemented since 2021. FMODB has been primarily constructed through the activities of the FMO drug design consortium (FMODD) [8,9]. As of this release, FMODB contains 77,277 FMO calculation entries covering 32,023 unique PDB IDs (Figure 1). It is increasingly being utilized as a foundational dataset for structural biology and drug discovery research because the FMO method enables the computation of inter-fragment interaction energies (IFIEs), which quantify the strength of interactions between small fragments such as amino acid residues, nucleic acids, ligands, solvent molecules, sugar chains, and metals [6,10–20]. Furthermore, IFIEs can be decomposed using pair interaction energy decomposition analysis (PIEDA) into physically meaningful components such as electrostatics, exchange repulsion, charge transfer, and dispersion [21,22]. These interaction energy descriptors form the core of FMODB. Thus, IFIE/PIEDA analysis enables precise and quantitative characterization of key interactions involved in molecular recognition, such as hydrogen bonds, ionic interactions, CH/π interactions, and π–π interactions [6,12,18,23]. In this report, we present an overview and technical background of the major new features recently implemented in FMODB, in the form of a release note. These developments were driven by requests from FMODD consortium members and FMODB tutorial participants [11,16,24–33]. For the features introduced in this release note, we have prepared a manual (https://drugdesign.riken.jp/FMODB/manual.php) and tutorial materials (in Japanese only) [28–33].

Figure 1. Annual trends in FMODB data registration by structure type and unique PDB IDs

2. Overview of new features

Table 1 provides a summary of the core features implemented in FMODB up to 2021, together with the new features added thereafter. As shown in Figure 2, although the FMODB home page provides access to various functions, only an Advanced Search interface is highlighted here as an example of the newly implemented features. Detailed descriptions and practical examples of each feature are provided in the following subsections.

Table 1. Overview of Core and Newly Implemented Features in FMODB

The rightmost column (“Detail description”) indicates the sections where each new feature is described in detail.

Feature Category	aCore Features	bNewly Implemented Features	Detail description
Search Functionality	ID search    Keyword search    Sequence-based BLAST search	Ligand structure search    Advanced search (filtering by ligand presence, FMO conditions, publication info, etc.)	Section 2.1.
IFIE Analysis for Individual Entries	IFIE/PIEDA analysis (one-to-one fragment pairs)    IFIE-MAP	Multi-fragment IFIE/PIEDA analysis (N-to-1 comparison)    IFIE diagram for interaction network visualization	Section 2.2.
Batch Analysis Support	Not available	Batch IFIE analysis for multiple FMO datasets (e.g., MD, docking, NMR)	Section 2.3.
Data Download	Entry-wise data download	Download multiple data files by specifying a list of FMODB IDs	Section 2.4.
Web API	Not available	Public Web API for programmatic data access	Section 2.5.
External Integration	Not available	Reciprocal cross-linking with PDBj    Integration with BioStation Viewer	Section 2.6.
Support for FMO Program	ABINIT-MP [2,34]	GAMESS [4,35]	Section 2.7.

^aFunctionality that had been implemented in FMODB prior to 2021.

^bNew functionality introduced in FMODB from 2021 onward.

2.1. Implementation of advanced search functionality

Conventional search functions in FMODB were limited to single-condition queries, such as searching by biological keywords or IDs (e.g., PDB, UniProt, and ChEMBL), or by sequence-based BLAST searches. The original interface lacked the flexibility to perform multi-criteria filtering, such as narrowing datasets based on structure information, FMO calculation parameters, or ligand presence. To address this limitation, we implemented a new “Advanced search” function (Figure 2; https://drugdesign.riken.jp/FMODB/advanced_search.php), which enables multifaceted queries. In addition to the existing search parameters, users can now filter data based on structure preprocessing methods, FMO calculation conditions, ligand presence or absence, and associated publication information. Furthermore, “Ligand structure search” capabilities—including both similarity and substructure searches—were introduced. These enhancements significantly improve accessibility and efficiency for ligand-centered data exploration within FMODB.

Figure 2. Web interface of FMODB (as of November 18, 2025)

(A) The current home page provides access to multiple functions of FMODB. For illustration, the entry point to the advanced search function is highlighted. (B) The advanced search panel enables multi-criteria queries such as filtering by ligand presence, calculation conditions, and publication information. (C) Example of the newly implemented IFIE diagram visualization for FMODB ID 7J1L3Z (PDB ID: 7BV2), showing the interaction energy network between the ligand, Remdesivir, and the surrounding amino acid residues and nucleic acid bases in the SARS-CoV-2 RNA-dependent RNA polymerase complex. Red and green lines indicate electrostatic and dispersion interactions, respectively, and numerical values represent IFIEs (in kcal/mol).

2.2. Enhancement of IFIE analysis capabilities

The original IFIE analysis in FMODB supported visualization of PIEDA results using stacked bar graphs for one-to-one fragment pairs (i.e., the base fragment of PIEDA/IFIE is a single fragment), such as interactions between amino acid residues or between an amino acid and a ligand. To extend this functionality, we newly implemented a “Multi-fragment IFIE/PIEDA analysis” feature that enables N-to-1 comparisons (i.e., the base fragments of PIEDA/IFIE are multiple fragments). For example, users can now analyze the cumulative interaction of N selected fragments (e.g., residues on an antigen) against each individual fragment (e.g., residues on an antibody), thereby facilitating the identification of binding hotspots [12,25]. In addition, we developed a novel visualization tool called the “IFIE diagram”, designed to intuitively visualize complex interaction networks [26,36,37]. This diagrammatic view enables users to explore IFIEs as a 2D schematic with customizable color schemes and display settings based on fragment type (e.g., amino acid, nucleic acid, ligand) and interaction type. This visualization aids in understanding the overall interaction landscape in a concise and accessible format. An example of the newly implemented IFIE diagram is shown in Figure 2C. This visualization represents the interaction energy network, as characterized by IFIE and PIEDA, between the ligand, Remdesivir, and the surrounding amino acid residues and nucleic acid bases within the SARS-CoV-2 RNA-dependent RNA polymerase complex. It highlights the strength and nature of each interaction.

2.3. Development of batch IFIE analysis functionality

We implemented a “Batch IFIE analysis” function that allows the simultaneous execution of IFIE and PIEDA calculations across multiple FMO datasets, provided that the protein structures and corresponding fragment definitions are consistent among them. This function is designed to handle sets of structures such as MD simulation snapshots, NMR ensembles, or docking poses. It enables efficient computation of dynamic averages of IFIEs and quantitative analysis of temporal variations in interaction energies, thereby facilitating integrated simulation–FMO studies, such as FMO combined with MD snapshots (FMO+MD) [27]. For docking datasets, batch analysis also supports extraction of protein–ligand interaction energies for structure–activity relationship (SAR) analyses.

2.4. Enhancement of data download functionality

Previously, users could only download data on a per-entry basis via the detail page of each FMODB entry. With the newly implemented feature, users can specify up to 50 FMODB IDs and download the corresponding datasets in bulk from “Download multiple data files” (https://drugdesign.riken.jp/FMODB/download.php). This enhancement greatly improves user convenience and facilitates more efficient large-scale data collection and batch analyses. For example, it smoothly supports visualized cluster analysis (VISCANA) [24,38], which classifies binding modes by clustering ligand interactions using FMO data from numerous protein–ligand complexes.

2.5. Provision of a Web API

To enable integration with external applications, as well as user-developed scripts and analysis tools, we released a “Web API” for FMODB (https://drugdesign.riken.jp/redoc/index.html). By specifying an FMODB ID, users can retrieve IFIE and PIEDA data in JSON format, allowing seamless incorporation of FMODB with other computational pipelines and custom workflows. The detailed API specification and usage examples are provided in the online documentation (https://drugdesign.riken.jp/redoc/index.html).

2.6. External integration

Leveraging the Web API, reciprocal cross-links were established between FMODB and structural information pages of PDB entries in PDBj [39–42] through the external database list (https://drugdesign.riken.jp/FMODB/pdbj.php), enabling bidirectional navigation between the two databases. This integration enhances accessibility to FMODB for researchers in the field of structural biology and promotes unified use of related databases. Furthermore, a direct connection feature to FMODB data was implemented in BioStation Viewer [43], a GUI tool designed specifically for FMO calculations. This integration allows users to seamlessly retrieve, visualize, and analyze FMODB data directly within the software environment, improving user convenience and promoting deeper data exploration.

2.7. Support for Additional FMO Program

Originally, FMODB contained only FMO calculation data obtained with ABINIT-MP [2,34]. Recently, support was extended to include FMO data calculated by GAMESS [4,35]. Specifically, a mechanism was implemented to generate check point files (CPFs) directly from GAMESS log files, enabling registration of GAMESS-derived FMO data into FMODB. This expansion increases the versatility of FMODB by allowing data from multiple FMO software packages to be archived and shared, thereby broadening its applicability across different computational environments.

3. Conclusion

In this release note, we summarized the major new features implemented in FMODB, together with their technical background, in the form of a release note. These improvements have significantly enhanced the searchability, analytical power, and interoperability of FMO data, thereby reinforcing FMODB as a fundamental resource for both academic and applied research. Future publications will present more detailed specifications and application examples of each function.

Acknowledgments

FMODB has been developed and released as a product of the FMO drug design consortium (FMODD). The authors sincerely thank all members of the FMODD consortium for their contributions of data and valuable technical input. We are also grateful to Prof. Yuji Mochizuki (Rikkyo University), Dr. Tatsuya Nakano (The Research Organization for Information Science and Technology), and Dr. Koji Okuwaki (JSOL Corporation) for providing ABINIT-MP, as well as to Dr. Dmitri G. Fedorov (National Institute of Advanced Industrial Science and Technology) and Dr. Alexander Heifetz (Sygnature Discovery) for providing GAMESS data and for their fruitful discussions on FMODB. We further acknowledge Prof. Genji Kurisu and Dr. Gert-Jan Bekker (Institute for Protein Research, the University of Osaka) for their insightful discussions and technical support in integrating of FMODB with PDBj. Finally, we thank Dr. Yusuke Kawashima and Dr. Yuma Handa (Hoshi University), Mr. Yuya Seki (TechnoPro, Inc.), Dr. Kazuki Watanabe (Chiba University), Mr. Shuhei Miyakawa and Ms. Ruri Mihata (Osaka University), and Dr. Yoshiharu Mori and Mr. Hiromu Matsumoto (Kyushu University) for their contributions as FMODB tutorial lecturers. This work was supported by 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 by JSPS KAKENHI Grant Numbers JP18K06619 and JP23K18192. FMO calculations were performed using the supercomputing resources of Fugaku (HPCI IDs: hp250154 and ra250009), HOKUSAI (Project ID: RB230116), TSUBAME (Project ID: ra000017), and SQUID (Project ID: hp240114).

References

• [1]  Kitaura, K.; Ikeo, E.; Asada, T.; Nakano, T.; Uebayasi, M. Fragment Molecular Orbital Method: An Approximate Computational Method for Large Molecules. Chem. Phys. Lett. 1999, 313 (3–4), 701–706.doi:10.1016/S0009-2614(99)00874-
• [2]  Tanaka, S.; Mochizuki, Y.; Komeiji, Y.; Okiyama, Y.; Fukuzawa, K. Electron-Correlated Fragment-Molecular-Orbital Calculations for Biomolecular and Nano Systems. Phys Chem Chem Phys 2014, 16 (22), 10310–10344. doi:10.1039/C4CP00316K.
• [3]  Recent Advances of the Fragment Molecular Orbital Method: Enhanced Performance and Applicability; Mochizuki, Y., Tanaka, S., Fukuzawa, K., Eds.; Springer: Singapore, 2021. doi:10.1007/978-981-15-9235-5.
• [4]  Fedorov, D. G. Complete Guide To The Fragment Molecular Orbital Method In Gamess: From One Atom To A Million, At Your Service; World Scientific, 2023.
• [5]  FMO DATABASE (FMODB). https://drugdesign.riken.jp/FMODB/ (accessed 2025-11-18).
• [6]  Takaya, D.; Watanabe, C.; Nagase, S.; Kamisaka, K.; Okiyama, Y.;et al. . FMODB: The World’s First Database of Quantum Mechanical Calculations for Biomacromolecules Based on the Fragment Molecular Orbital Method. J. Chem. Inf. Model. 2021,61(2),777–794 doi:10.1021/acs.jcim.0c01062.
• [7]  Watanabe, C.; Watanabe, H.; Okiyama, Y.; Takaya, D.; Fukuzawa, K.; et al. Development of an Automated Fragment Molecular Orbital (FMO) Calculation Protocol toward Construction of Quantum Mechanical Calculation Database for Large Biomolecules . CBIJ. 2019, 19 (0), 5–18. doi:10.1273/cbij.19.5.
• [8]  Watanabe, C.; Watanabe, H.; Okiyama, Y.; Takaya, D.; Fukuzawa, K.; et al. Development of an Automated Fragment Molecular Orbital (FMO) Calculation Protocol toward Construction of Quantum Mechanical Calculation Database for Large Biomolecules . CBIJ. 2019, 19 (0), 5–18. doi:10.1273/cbij.19.5.
• [9]  FMO Drug Design Consortium (FMODD). https://fmodd.jp. doi:10.1007/978-981-15-9235-58.
• [10]  Iwasaki, S.; Iwasaki, W.; Takahashi, M.; Sakamoto, A.; Watanabe, C.; et al.; Floor, S. N.; Fujiwara, K.; Mito, M.; Dodo, K.; Sodeoka, M.; Imataka, H.; Honma, T.; Fukuzawa, K.; Ito, T.; Ingolia, N. T. The Translation Inhibitor Rocaglamide Targets a Bimolecular Cavity between eIF4A and Polypurine RNA. Mol Cell. 2019, 73 (4), 738-748.e9. doi:10.1016/j.molcel.2018.11.026.
• [11]  Fukuzawa, K.; Kato, K.; Watanabe, C.; Kawashima, Y.; Handa, Y.; et al. Special Features of COVID-19 in the FMODB: Fragment Molecular Orbital Calculations and Interaction Energy Analysis of SARS-CoV-2-Related Proteins. J. Chem. Inf. Model. 2021, 61 (9), 4594–4612. doi:10.1021/acs.jcim.1c00694.
• [12]  Watanabe, C.; Okiyama, Y.; Tanaka, S.; Fukuzawa, K.; Honma, T. Molecular Recognition of SARS-CoV-2 Spike Glycoprotein: Quantum Chemical Hot Spot and Epitope Analyses. Chem. Sci. 2021, 12 (13), 4722–4739. doi:10.1039/D0SC06528E.Imai,
• [13]  K.; Takimoto, D.; Saito, R.; Watanabe, C.; Fukuzawa, K.; et al. FMO Calculations for Zinc Metalloprotease:Fragmentation of Amino-Acid Residues Coordinated to Zinc Ion. CBIJ. 2022, 22 (0), 21–25. doi:10.1273/cbij.22.21.
• [14]  Fujii, M.; Watanabe, C.; Fukuzawa, K.; Tanaka, S. Fragment Molecular Orbital Calculations Containing Mg²⁺ Ions: PPlase Domain of Cyclophilin G. CBIJ. 2022, 22 (0), 55–62. doi.org:10.1273/cbij.22.55.
• [15]  Sengoku, T.; Shiina, M.; Suzuki, K.; Hamada, K.; Sato, K.; et al. Structural Basis of Transcription Regulation by CNC Family Transcription Factor, Nrf2. Nucleic Acids Res. 2022, 50 (21), 12543–12557. doi:10.1093/nar/gkac1102.
• [16]  Watanabe, C.; Tanaka, S.; Okiyama, Y.; Yuki, H.; Ohyama, T.; et al. Quantum Chemical Interaction Analysis between SARS-CoV-2 Main Protease and Ensitrelvir Compared with Its Initial Screening Hit. J. Phys. Chem. Lett. 2023, 14 (15), 3609–3620. doi.:10.1021/acs.jpclett.2c03768.
• [17]  Kato, K.; Yamamoto, A.; Watanabe, C.; Fukuzawa, K. Application of Model Core Potentials to Zn- and Mg-Containing Metalloproteins in the Fragment Molecular Orbital Method. CBIJ. 2023, 23 (0), 14–25. doi:10.1273/cbij.23.14.
• [18]  Kamisaka, K.; Nagase, S.; Watanabe, C.; Takaya, D.; Yuki, H.; et al. Statistical Analysis of Interactions among Amino Acid Residues in Apo Structures Using Fragment Molecular Orbital Method. CBIJ. 2024, 24 (0), 25–47. doi:10.1273/cbij.24.25
• [19]  Moriwaki, H.; Kawashima, Y.; Watanabe, C.; Kamisaka, K.; Okiyama, Y.; et al. FMOe: Preprocessing and Visualizing Package of the Fragment Molecular Orbital Method for Molecular Operating Environment and Its Applications in Covalent Ligand and Metalloprotein Analyses. J. Chem. Inf. Model. 2024, 64 (18), 6927–6937. doi:10.1021/acs.jcim.4c01169.
• [20]  Takaya, D.; Ohno, S.; Miyagishi, T.; Tanaka, S.; Okuwaki, K.; et al. Quantum Chemical Calculation Dataset for Representative Protein Folds by the Fragment Molecular Orbital Method. Sci. Data 2024, 11 (1), 1164. doi:10.1038/s41597-024-03999-2.
• [21]  Fedorov, D. G.; Kitaura, K. Pair Interaction Energy Decomposition Analysis. J. Comput. Chem. 2007, 28 (1), 222–237. doi:10.1002/jcc.20496.
• [22]  Tsukamoto T.; Kato K.; Kato A.; Nakano T.; Mochizuki Y.; et al. Implementation of Pair Interaction Energy DecompositionAnalysis and Its Applications to Protein-Ligand Systems. J. Comput. Chem. Jpn. 2015, 14 (1), 1–9. doi:10.2477/jccj.2014-0039.
• [23]  Ozawa, T.; Okazaki, K.; Kitaura, K. CH/π Hydrogen Bonds Play a Role in Ligand Recognition and Equilibrium between Active and Inactive States of the Β2 Adrenergic Receptor: An Ab Initio Fragment Molecular Orbital (FMO) Study. Bioorg. Med. Chem. 2011, 19 (17), 5231–5237. doi:10.1016/j.bmc.2011.07.004.
• [24]  Seki, Y.; Kato, T.; Furuishi, T.; Fukuzawa, K.; Yonemochi, E. Analysis of Ligand Binding Specificity of Estrogen Receptor by the Fragment Molecular Orbital Method. J. Comput. Chem. Jpn. 2018, 17 (3), 160–162. doi:10.2477/jccj.2018-0028.
• [25]  Watanabe, K.; Watanabe, C.; Honma, T.; Tian, Y.-S.; Kawashima, Y.; et al. Intermolecular Interaction Analyses on SARS-CoV-2 Spike Protein Receptor Binding Domain and Human Angiotensin-Converting Enzyme 2 Receptor-Blocking Antibody/Peptide Using Fragment Molecular Orbital Calculation. J. Phys. Chem. Lett. 2021, 12 (16), 4059–4066. doi:10.1021/acs.jpclett.1c00663.
• [26]  Kato, K.; Honma, T.; Fukuzawa, K. Intermolecular Interaction among Remdesivir, RNA and RNA-Dependent RNA Polymerase of SARS-CoV-2 Analyzed by Fragment Molecular Orbital Calculation. J. Mol. Graph. Model. 2020, 100, 107695. doi:10.1016/j.jmgm.2020.107695.
• [27]  Takaba, K.; Watanabe, C.; Tokuhisa, A.; Akinaga, Y.; Ma, B.; et al. Protein–Ligand Binding Affinity Prediction of Cyclin‐dependent Kinase‐2 Inhibitors by Dynamically Averaged Fragment Molecular Orbital‐based Interaction Energy. J. Comput. Chem. 2022, 43 (20), 1362–1371. doi:10.1002/jcc.26940.
• [28]  Fukuzawa, K.; Watanabe, C.; Honma, T.; Takaya, D.; Kamisaka, K.; et al. FMODB Tutorial 2020 (Japanese Documentation), 2020. https://fmodd.jp/biostationviewer-dl/ (accessed 2025-11-18).
• [29]  Fukuzawa, K.; Takaya, D.; Handa, Y.; Watanabe, K.; Watanabe, C.; et al. FMODB Tutorial 2021 (Japanese Documentation), 2021. https://drugdesign.riken.jp/pub/CBI2021tut/ (accessed 2025-11-18).
• [30]  Kato, K.; Takaya, D.; Watanabe, C.; Kamisaka, K.; Miyakawa, S.; et al. FMODB Tutorial 2022 (Japanese Documentation), 2022. https://drugdesign.riken.jp/pub/CBI2022tut/ (accessed 2025-11-18).
• [31]  Takaya, D.; Watanabe, C.; Kato, K.; Matsumoto, H.; Kamisaka, K. FMODB Tutorial 2023 (Japanese Documentation), 2023. https://drugdesign.riken.jp/pub/CBI2023tut/ (accessed 2025-11-18).
• [32]  Watanabe, C.; Kato, K.; Miyakawa, S.; Mihata, R.; Kamisaka, K.; et al. FMODB Tutorial 2024 (Japanese Documentation), 2024. https://drugdesign.riken.jp/pub/CBI2024tut/ (accessed 2025-11-18).
• [33]  Kato, K.; Takaya, D.; Ohyama, T.; Mori, Y.; Kamisaka, K.; et al FMODB Tutorial 2025 (Japanese Documentation), 2025. https://drugdesign.riken.jp/pub/CBI2025tut/ (accessed 2025-11-18).
• [34]  Mochizuki, Y.; Nakano, T.; Sakakura, K.; Okiyama, Y.; Watanabe, H.; et al. The ABINIT-MP Program. Recent Advances of the Fragment Molecular Orbital Method: Enhanced Performance and Applicability, 2021, 53–67. doi.:10.1007/978-981-15-9235-5_4.
• [35]  Gordon Group/GAMESS Homepage. https://www.msg.chem.iastate.edu/gamess/index.html (accessed 2025-11-18).
• [36]  Fukuzawa, K.; Watanabe, C.; Kurisaki, I.; Taguchi, N.; Mochizuki, Y.; et al. Accuracy of the Fragment Molecular Orbital (FMO) Calculations for DNA: Total Energy, Molecular Orbital, and Inter-Fragment Interaction Energy. Comput. Theor. Chem. 2014, 1034, 7–16. doi:10.1016/j.comptc.2014.02.002.
• [37]  Fukuzawa, K.; Kurisaki, I.; Watanabe, C.; Okiyama, Y.; Mochizuki, Y.; et al. Explicit Solvation Modulates Intra- and Inter-Molecular Interactions within DNA: Electronic Aspects Revealed by the Ab Initio Fragment Molecular Orbital (FMO) Method. Comput. Theor. Chem. 2015, 1054, 29–37. doi:10.1016/j.comptc.2014.11.020.
• [38]  Amari, S.; Aizawa, M.; Zhang, J.; Fukuzawa, K.; Mochizuki, Y.; et al. VISCANA: Visualized Cluster Analysis of Protein−Ligand Interaction Based on the Ab Initio Fragment Molecular Orbital Method for Virtual Ligand Screening. J. Chem. Inf. Model. 2006, 46 (1), 221–230. doi.:10.1021/ci050262q.
• [39]  Protein Data Bank Japan (PDBj). https://pdbj.org (accessed 2025-11-18).
• [40]  Kinjo, A. R.; Bekker, G.-J.; Suzuki, H.; Tsuchiya, Y.; Kawabata, T.; et al. Protein Data Bank Japan (PDBj): Updated User Interfaces, Resource Description Framework, Analysis Tools for Large Structures. Nucleic Acids Res. 2017, 45 (D1), D282–D288. doi:10.1093/nar/gkw962.
• [41]  Kinjo, A. R.; Bekker, G.; Wako, H.; Endo, S.; Tsuchiya, Y.; et al. New Tools and Functions in Data‐out Activities at Protein Data Bank Japan (PDBj). Protein Sci. 2018, 27 (1), 95–102. doi:10.1002/pro.3273.
• [42]  Bekker, G.; Yokochi, M.; Suzuki, H.; Ikegawa, Y.; Iwata, T.; et al. Protein Data Bank Japan: Celebrating Our 20th Anniversary during a Global Pandemic as the Asian Hub of Three Dimensional Macromolecular Structural Data. Protein Sci. 2022, 31 (1), 173–186. doi:10.1002/pro.4211.
• [43]  Kato, A. BioStation Viewer. https://fmodd.jp/biostationviewer-dl/ (accessed 2025-11-18).