Fluorescence — and, in particular, Förster resonance energy transfer (FRET) — experiments provide rich insights into biomolecular systems, including structural and dynamic information, and are increasingly used for integrative structure modeling of biomolecules1,2,3 (see Supplementary Note 1 for an overview).

In integrative structure modeling approaches, data from different experimental and theoretical techniques are combined to maximally utilize method-specific advantages and complementary information (Supplementary Note 2)4. Making these results publicly available following the FAIR (findable, accessible, interoperable, reusable) principles — for example, in structural biology model repositories (Supplementary Note 3) — is crucial for advancing science5,6. The prototype PDB-Dev repository (https://pdb-dev.wwpdb.org/)7,8, which was recently unified with the PDB archive, offers a framework for deposition, curation, validation, archiving and dissemination of integrative structure models. PDB-Dev uses the Protein Data Bank Exchange / macromolecular Crystallographic Information Framework (PDBx/mmCIF) data standard9, along with an extension developed for representing Integrative and Hybrid Models (IHMCIF; https://github.com/ihmwg/IHMCIF)7,10 (see Supplementary Note 4 for details on the dictionaries). IHMCIF is supported by the python-ihm library (https://github.com/ihmwg/python-ihm), which can be used to read and write IHMCIF-compliant files for PDB-Dev, and provides an application programming interface (API) for other software.

To support the deposition and validation of fluorescence-based integrative structure models in PDB-Dev, we present flrCIF (https://github.com/ihmwg/flrCIF), an extension of PDBx/mmCIF and IHMCIF. flrCIF was developed as a method-specific extension dictionary that allows seamless interoperation of fluorescence and FRET data with PDB and PDB-Dev data. Together with PDBx/mmCIF and IHMCIF, flrCIF enables deposition of structure models, restraint data, references to experimental raw data in other repositories, and additional metadata that support FAIR principles (Fig. 1a; see Supplementary Note 5 for data definitions).

Fig. 1: Information in dictionary-compliant file for a FRET-assisted integrative structure model.
figure 1

a, Information content of the different dictionaries supporting deposition of FRET-based structure models in PDB-Dev. PDBx/mmCIF describes general information about the entry, such as description of the molecular system and atomic coordinates as well as metadata regarding authors, software and citations. IHMCIF contains definitions related to integrative modeling and descriptions of multi-scale, multi-state and ordered models and collections of models. flrCIF provides definitions for fluorescence-experiment-related information as shown in b. The flr2mmcif converter script uses a spreadsheet to collect information in a human-readable format and converts this to a PDB-Dev-compatible IHMCIF file using the python-ihm library (Supplementary Note 8). b, Data definitions for fluorescence and FRET-assisted structure modeling. Components of the flrCIF dictionary complement the PDBx/mmCIF and IHMCIF dictionaries for the description of fluorescence-related information. The green boxes show definitions in PDBx/mmCIF; blue boxes, definitions in IHMCIF; yellow boxes, definitions in flrCIF.

Full size image

In the current version, flrCIF covers four key workflow aspects from the fluorescence experiment to the structure model (Fig. 1b and Supplementary Note 6): (1) experiment and sample description including setup, conditions, and fluorescent probes with their attachment; (2) analysis workflow with correction parameters and reference measurements; (3) structure modeling procedure with FRET-derived distance restraints and modeling software settings and parameters; and (4) assessment of structure models on the basis of input experimental data.

While the current version of flrCIF is focused on FRET experiments, the dictionary is extensible and can be further expanded to include descriptions of other fluorescence experiments, such as fluorescence quenching assays or fluorescence anisotropy measurements used to probe the local dynamics, as well as other analysis approaches (see Supplementary Note 6 for extension plans).

In addition to creating fluorescence-specific definitions in flrCIF, we also extended the IHMCIF dictionary with generic definitions applicable to other probe-based methods. Importantly, a model in IHMCIF can be multi-state10 (see Supplementary Note 7 for details on molecular kinetics and multi-state models). This allowed us to introduce descriptions of complex kinetic schemes connecting multiple states, exchange kinetics, and intrastate dynamics characterizing flexibility. To quantify the kinetics, we included characteristic observables such as equilibrium constants, population fractions, relaxation times and transition rate constants (see Supplementary Note 7 for details and an example). The deposition of fluorescence-aided structure models in PDB-Dev is outlined in Fig. 1a. PDB-Dev deposition requires an mmCIF file compliant with flrCIF, as well as the parent PDBx/mmCIF and IHMCIF dictionaries. To facilitate the preparation of such a file, we extended the python-ihm library to support the new definitions in flrCIF (Supplementary Note 8).

To facilitate the submission of experimental information that may be conveniently collected in a spreadsheet (see Fig. 1 for content categories, with more details in Supplementary Note 8), we also created a converter script (flr2mmcif), which uses the python-ihm library, to convert the data captured in a spreadsheet to flrCIF and IHMCIF-compliant files for submission to PDB-Dev (https://github.com/Fluorescence-Tools/flr2mmcif).

The development of flrCIF offers possibilities for different interconnected user groups and applications:

1. Researchers and data users:

  • For scientists performing integrative structure modeling using fluorescence data, flrCIF provides a standardized framework and tools for deposition of fluorescence-assisted models in PDB-Dev (examples in Supplementary Note 9).

  • Data consumers benefit from standardized storage of fluorescence-based structure models and information, facilitating access and reuse in further studies.

  • Both groups profit from links to metadata, protocols and experimental data in other repositories through existing definitions in PDBx/mmCIF and IHMCIF.

2. Software developers:

  • flrCIF and the extended python-ihm library provide a convenient framework for handling fluorescence-related information in software applications.

  • Access to structure models alongside experimental data supports the development and validation of new computational modeling and analysis approaches.

3. Broader scientific applications:

  • flrCIF establishes data definitions adaptable to other experimental methods, such as electron paramagnetic resonance spectroscopy, particularly descriptions for probe chemistry and attachment.

  • The inclusion of complex kinetic networks and intra-state dynamics in IHMCIF extends the applicability beyond fluorescence to various techniques used in integrative modeling studies and facilitates the collection and archiving of structural dynamics information.

Overall, flrCIF contributes toward dynamic structural biology by enabling standardized deposition, archiving and dissemination of information on biomolecules in their functional context — from static structures toward dynamic structural networks and the connection of structures with energy landscapes and real time motions.

Reporting Summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.