The structures of many complex macromolecular assembles are obtained through integrative/hybrid (I/H) methods. These methods involve a combination of complementary methods, both experimental and computational, that provide information on different aspects of the assembly that can be integrated to obtain a structural view of the complex biological machinery. The models obtained through integrative/hybrid methods are more complex than the traditional structural models archived in the Protein Data Bank (PDB) that are obtained from a single method such as X-ray crystallography, NMR spectroscopy or electron microscopy. We aim to create an archive for integrative/hybrid methods and propose to extend the existing PDBX/mmCIF dictionary to accommodate the requirements for integrative models.
Figure 1. Integrative modeling of a complex macromolecular assembly utilizes spatial restraints dervied from a variety of experiments along with available structural data of individual components to drive an extensive computational sampling and analysis leading to the structural characterization of the assembly.
The PDBX/mmCIF dictionary is currently used by the Protein Data Bank to archive structural models of biologcal macromolecules. The details of the dictionary and the corresponding documentataion can be obtained from the wwpdb mmCIF dictionary website. The documention on the dictionary file syntax, dictionary organization, atomic description and molecular description are especially recommended.
An extension to the existing PDBX/mmCIF dictionary has been created to capture the complex details involved in generating integrative/hybrid models. The figure below provides a high-level illustration of the content areas incorporated in the I/H methods extension dictionary.
Figure 2. Illustration of the data content captured in the IHMCIF dictionary extension.
The I/H methods extension dictionary describes the different kinds of input data types used in integrative modeling. These include the following.
-
Description of the molecular components in the macromolecular assembly such as proteins, nucleic acids and ligands. The primary sequence of proteins and nucleic acids, the identities and chemical composition of the ligands and the number of copies of each of these molecular components, follow from the corresponding definitions in the PDBx/mmCIF dictionary. These data items are obtained from the respective sequence and chemical repositories such as UniProt, GenBank and the Cambridge Crystallographic Data Center, CCDC and are referenced accordingly in the dictionary.
-
The starting structural models of individual molecular components can be obtained from experimental methods such as X-ray crystallography and NMR spectroscopy or from insilico methods such as comparative and abinitio modeling. These starting models are obtained from the respective structural data repositories such as the Protein Data Bank, PDB and the model archive at the Protein model portal, PMP, and are referenced accordingly in the dictionary.
-
The input spatial restraints used in integrative modeling can be derived from various experimental methods such as electron microscopy (EM), nuclear magnetic resonance spectroscopy (NMR), cross-linking mass spectrometry (CX-MS), small angle solution scattering (SAS), Forster resonance energy transfer (FRET), atomic force microscopy (AFM), electron paramagnetic resonance spectroscopy (EPR) and various other proteomics methods such as affinity purification, co-purification etc. These categories further account for the ambiguities and uncertainties involved in the interpretation of experimental data and the corresponding spatial restraints. References are included for the primary data archived in the corresponding external repositories such as the PDB, BMRB, EMDB, EMPIAR, SASBDB and PRIDE. Allowances have been made to include data that do not yet have a primary repository. These can be linked via persistent digital object identifiers (DOI).
The integrative modeling algorithm can involve any combination of the following and are defined in the I/H methods dictionary.
- Multi-scale representation of macromolecules including atomistic and non-atomistic representations such as spheres/beads, three-dimensional Gaussians.
- Multiple conformational states of the macromolecular assembly such as open and closed states or ligand bound and unbound states.
- A time-ordered ensemble representing a trajectory, reaction cycle or metabolic pathway.
The dictionary also includes definitions for the integrative modeling workflow, which can be complex involving multiple stages of sampling, scoring and post-processing (such as filtering and clustering). The structural details of the ensembles of macromolecular assemblies obtained through integrative modeling are also captured in the I/H methods dictionary.
The I/H methods dictionary consists of multiple data categories that are designed to capture the information detailed above. Some examples are elucidated in the sections below.
Integrative modeling encompasses a multitude of experimental data types that carry
complementary information about the macromolecular assembly. Spatial restraints
derived from various experiments are used together to obtain the structure of the
macromolecular assembly. All input restraints derived from experiments are listed
in the ihm_dataset_list data category. These can be grouped together to distinguish
data sets used differently in the integrative modeling. The ihm_related_datasets
category holds the relationships between datasets derived from primary datasets.
Some experimental methods such as NMR spectroscopy (BMRB),
Electron microscopy (EMDB, EMPIAR),
Small angle solution scattering SASBDB), crosslinking mass spectrometry
and other proteomics data (PRIDE) have their own repositories
where primary data is hosted. Data derived from such sources are listed appropriately with references
pointing to the primary data resource. The details of database name, accession code, version number are
included in the ihm_dataset_related_db_reference data category.
Other data types that do not currently have a primary repository can be included via persistent
digital object identifiers (DOIs). The DOIs and the paths to the related external files are included
in the ihm_external_reference_info and the ihm_external_files categories. These are linked to the
input datasets in ihm_dataset_list via the ihm_dataset_external_reference category. DOIs can be obtained
for individual files or entire datasets (inputs, outputs, scripts etc.) hosted at the author wesbite
or other websites such as Github. DOIs for such datasets can be obtained from
resources such as Figshare or Zenodo.
The various kinds of input data used in the integrative modeling are listed and described in the ihm_dataset_list,
ihm_related_datasets, ihm_dataset_related_db_reference, ihm_external_reference_info, ihm_external_files and
ihm_dataset_external_reference categories (examples below). The restraints obtained from these data sources are
handled in other data categories discussed in the following section. The ihm_external_reference_info and the
ihm_external_files are generic categories that link out to information in standard formats stored in external files
referenced via a DOI or as supplementary files stored locally. These are especially useful for referencing data
currently not archived in other databases.
Example for ihm_dataset_list category
loop_
_ihm_dataset_list.ordinal_id
_ihm_dataset_list.id
_ihm_dataset_list.group_id
_ihm_dataset_list.data_type
_ihm_dataset_list.database_hosted
1 1 1 'CX-MS data' 'NO'
2 2 1 'CX-MS data' 'NO'
3 3 1 '3DEM volume' 'YES'
4 4 1 '2DEM class average' 'NO'
5 5 1 'Comparative model' 'NO'
6 6 1 'EM raw micrographs' 'NO'
# ... abbreviated ...
Example for ihm_related_datasets category
loop_
_ihm_related_datasets.ordinal_id
_ihm_related_datasets.dataset_list_id_derived
_ihm_related_datasets.dataset_list_id_primary
1 1 5
2 2 6
3 3 7
4 4 8
# ... abbreviated ... #
Example for ihm_dataset_related_db_reference category
loop_
_ihm_dataset_related_db_reference.id
_ihm_dataset_related_db_reference.dataset_list_id
_ihm_dataset_related_db_reference.db_name
_ihm_dataset_related_db_reference.accession_code
_ihm_dataset_related_db_reference.version
_ihm_dataset_related_db_reference.details
1 6 'PRIDE' 'PXD00381' . 'Structural Analysis of a Prokaryotic Ribosome Using a Novel Amidinating Cross-Linker and Mass Spectrometry'
2 3 'EMDB' 'EMD-2799' . .
3 5 'PDB' '5FM1' . 'STRUCTURE OF GAMMA-TUBULIN SMALL COMPLEX BASED ON A CRYO-EM MAP, CHEMICAL CROSS-LINKS, AND A REMOTELY RELATED STRUCTURE'
4 7 'BMRB' '25766' . .
5 8 'MODEL ARCHIVE' 'MA-CO2KC' . 'PREDICTED STRUCTURE OF AN ALPHA NEUROTOXIN FROM TAIWAN COBRA BY HOMOLOGY MODELING'
# ... abbreviated ...
Example for ihm_dataset_external_reference category
loop_
_ihm_dataset_external_reference.id
_ihm_dataset_external_reference.dataset_list_id
_ihm_dataset_external_reference.file_id
1 1 1
2 2 2
3 4 3
4 4 4
5 5 5
# ... abbreviated ...
Example 1 for ihm_external_reference_info category (DOI reference leading to an archive file)
loop_
_ihm_external_reference_info.reference_id
_ihm_external_reference_info.reference_provider
_ihm_external_reference_info.reference_type
_ihm_external_reference_info.reference
_ihm_external_reference_info.refers_to
_ihm_external_reference_info.associated_url
1 'Zenodo' 'DOI' '10.5281/zenodo.46266' 'Archive' 'https://zenodo.org/record/46266/files/nup84-v1.0.zip'
# ... abbreviated ...
Example 2 for ihm_external_reference_info category (Supplementaty files stored locally)
_ihm_external_reference_info.reference_id 2
_ihm_external_reference_info.reference_type 'Supplementary Files'
_ihm_external_reference_info.reference .
_ihm_external_reference_info.refers_to 'Other'
Example 3 for ihm_external_reference_info category (DOI reference leading to a single file)
loop_
_ihm_external_reference_info.reference_id
_ihm_external_reference_info.reference_provider
_ihm_external_reference_info.reference_type
_ihm_external_reference_info.reference
_ihm_external_reference_info.refers_to
_ihm_external_reference_info.associated_url
3 'Zenodo' 'DOI' '10.5281/zenodo.58025' 'File' 'https://zenodo.org/record/58025/files/Nup84complex_particles.spd'
# ... abbreviated ...
Example 1 for ihm_external_files category (DOI Archive Files)
loop_
_ihm_external_files.id
_ihm_external_files.reference_id
_ihm_external_files.file_path
_ihm_external_files.file_format
_ihm_external_files.content_type
_ihm_external_files.file_size_kb
_ihm_external_files.details
1 1 'integrativemodeling-nup84-a69f895/data/EDC_XL_122013.new.dat' 'TXT' 'Input data or restraints' 3.16 'EDC crosslinks file'
2 1 'integrativemodeling-nup84-a69f895/data/yeast_Nup84_DSS.new.dat' 'TXT' 'Input data or restraints' 4.11 'DSS crosslinks file'
3 1 'integrativemodeling-nup84-a69f895/data/nup84_kinked_from_class2.pgm' 'PGM' 'Input data or restraints' 43.1 'Nup84 2DEM class average'
4 1 'integrativemodeling-nup84-a69f895/data/nup84_kinked_from_class2.mrc' 'MRC' 'Input data or restraints' . 'Nup84 2DEM class average'
5 1 'integrativemodeling-nup84-a69f895/data/ScSec13_2-296_new2.pdb' 'PDB' 249.0 'Input data or restraints' 'Sec13 starting comparative model'
6 1 'integrativemodeling-nup84-a69f895/scripts/MODELLER_scripts/Nup84/all_align_final2.ali' 'PIR/NBRF' 'Input data or restraints' 1.6 'Nup84 multiple sequence alignment file'
7 1 'integrativemodeling-nup84-a69f895/scripts/MODELLER_scripts/Nup84/all_align_final1.ali' 'PIR/NBRF' 'Input data or restraints' 5.0 'Nup84 multiple sequence alignment file'
8 1 'integrativemodeling-nup84-a69f895/scripts/nup84.isd.modeling.py' 'Other' 'Modeling workflow or script' 6.8 'Python integrative modeling main script'
9 1 'integrativemodeling-nup84-a69f895/scripts/nup84.topology.py' 'Other' 'Modeling workflow or script' 5.6 'Python script for constructing subunits'
10 1 'integrativemodeling-nup84-a69f895/scripts/nup84.merge.py' 'Other' 'Modeling workflow or script' 1.5 'Python script to merge output files'
11 1 'integrativemodeling-nup84-a69f895/scripts/README' 'TXT' 'Modeling workflow or script' 5.7 'Readme file'
12 1 'integrativemodeling-nup84-a69f895/outputs/localization/cluster1/nup84.mrc' 'MRC' 'Modeling or post-processing output' . 'Nup84 localization density file'
13 1 'integrativemodeling-nup84-a69f895/outputs/localization/cluster1/nup85.mrc' 'MRC' 'Modeling or post-processing output' . 'Nup85 localization density file'
14 1 'integrativemodeling-nup84-a69f895/outputs/localization/cluster1/nup120.mrc' 'MRC' 'Modeling or post-processing output' . 'Nup120 localization density file'
# ... abbreviated ...
Example 2 for ihm_external_files category (Supplementary Local Files)
loop_
_ihm_external_files.id
_ihm_external_files.reference_id
_ihm_external_files.file_path
_ihm_external_files.file_format
_ihm_external_files.content_type
_ihm_external_files.file_size_kb
_ihm_external_files.details
1 2 'data/EDC_XL_122013.dat' 'TXT' 'Input data or restraints' 2.4 'EDC crosslinks file'
2 2 'data/yeast_Nup84_DSS.dat' 'TXT' 'Input data or restraints' 3.7 'DSS crosslinks file'
3 2 'nup145.mrc' 'MRC' 'Modeling or post-processing output' . 'Nup145 localization density file'
# ... abbreviated ...
Example 3 for ihm_external_files category (DOI reference leading to a single file)
loop_
_ihm_external_files.id
_ihm_external_files.reference_id
_ihm_external_files.file_path
_ihm_external_files.file_format
_ihm_external_files.content_type
_ihm_external_files.file_size_kb
_ihm_external_files.details
1 3 . 'Input data or restraints' 'Raw micrographs from which the class average was derived'
# ... abbreviated ...
Input restraints are obtained from various kinds of experimental data including electron microscopy, chemical crosslinking followed by mass spectrometry, small angle solution scattering, Forster resonance energy tranfer, electron paramagnetic resonance spectroscopy and other Proteomics experiments. The current I/H methods dictionary extension includes restraints obtained from 2DEM, 3DEM and CX-MS experiments. Restraints derived from other methods such as SAXS, EPR and FRET will soon be incorporated into the dictionary. Sometimes, the integrative model may correspond to only a subset of the experimental data (e.g. an EM map belonging to a complex is used to model a subcomplex) or vice-versa (e.g. the modeling involves the entire complex, however the EM data belongs to subcomplexes obtained via domain deletion). Such details are captured in the various categories that describe restraints derived from experimental data.
Chemical crosslinking experiments provide information on distances between amino acid residues involved in the crosslink. Crosslinks can occur within a protein molecule (intra-chain crosslinks) or between different protein molecules (inter-chain crosslinks). For example, residue i in Protein A may be crosslinked to residue j in Protein B or residues i and j from Protein A may be crosslinked to each other.
Figure 3. Chemical crosslinking experiments provide information regarding distances between the crosslinked amino acid residues. The crosslinks across subunits within an assembly can be very useful in integrative modeling.
Restraint data obtained from crosslinking experiments are captured in the
ihm_cross_link_list category. The crosslinks can be grouped together to
enumerate experimental ambiguities or uncertainties regarding residue
identities, where it is not clear whether residue i or i+2 is crosslinked
to residue j. In such cases both crosslinks (i to j and i+2 to j) are
enumerated in the crosslink list, but are grouped together to indicate
that they come from a single experimental observation. The type of
chemical crosslink used in the experiment and the input dataset id are
also provided.
Example for ihm_cross_link_list category
loop_
_ihm_cross_link_list.id
_ihm_cross_link_list.group_id
_ihm_cross_link_list.entity_description_1
_ihm_cross_link_list.entity_id_1
_ihm_cross_link_list.seq_id_1
_ihm_cross_link_list.comp_id_1
_ihm_cross_link_list.entity_description_2
_ihm_cross_link_list.entity_id_2
_ihm_cross_link_list.seq_id_2
_ihm_cross_link_list.comp_id_2
_ihm_cross_link_list.type
_ihm_cross_link_list.dataset_list_id
1 1 Nup120 3 17 LYS Nup120 3 421 GLU EDC 1
2 2 Nup120 3 206 LYS Nup120 3 219 ASP EDC 1
3 3 Nup120 3 38 LYS Nup120 3 429 GLU EDC 1
4 4 Nup120 3 400 LYS Nup120 3 440 GLU EDC 1
5 4 Nup120 3 400 LYS Nup120 3 441 GLU EDC 1
6 5 Nup120 3 400 LYS Nup120 3 523 ASP EDC 1
7 6 Nup120 3 735 LYS Nup120 3 403 ASP EDC 1
8 6 Nup120 3 735 LYS Nup120 3 405 GLU EDC 1
9 7 Nup120 3 865 LYS Nup120 3 894 GLU EDC 1
10 7 Nup120 3 865 LYS Nup120 3 898 ASP EDC 1
11 8 Nup133 4 103 LYS Nup133 4 130 ASP EDC 1
12 8 Nup133 4 103 LYS Nup133 4 131 GLU EDC 1
13 9 Nup120 3 17 LYS Nup120 3 412 LYS DSS 2
14 10 Nup120 3 17 LYS Nup120 3 735 LYS DSS 2
15 11 Nup120 3 206 LYS Nup120 3 217 LYS DSS 2
16 12 Nup120 3 384 LYS Nup120 3 362 LYS DSS 2
17 13 Nup120 3 400 LYS Nup120 3 530 LYS DSS 2
18 14 Nup120 3 400 LYS Nup120 3 531 LYS DSS 2
19 15 Nup120 3 400 LYS Nup120 3 735 LYS DSS 2
20 16 Nup120 3 735 LYS Nup120 3 412 LYS DSS 2
# ... abbreviated ...
Implementation of the crosslinking data as spatial restraints in the
integrative modeling is captured in the ihm_cross_link_restraint category.
This category captures the following details regarding crosslinking restraints.
- Multiple copies of Protein A or B may exist. In such cases, the crosslinks are grouped together to indicate multiple instances arising from copies of the same protein(s) exisiting in the assembly.
- The grouped restraints can be encoded as conditional crosslinks where any or all crosslinks can be required to be satisfied during modeling.
- Sometimes the residues involved in the crosslinks may be absent or may be represented at a lower resolution, such as part of a larger sphere or bead in multi-scale models. In such cases with crosslinking residues in coarse-grained representations, the model granularity information is included to indicate how the crosslinks are represented in the modeling algorithm.
- The distance threshold applied to the crosslink is provided. This is dependent on the type of the chemical moeity involved in the crosslink.
- Uncertainties in the crosslinking data can be included as non-cartesian
parameters or Bayesian nuisances
- The uncertainty in the crosslinking experiment data may be approximated to the false positive rate and is represented by the psi value in integrative modeling.
- The uncertainty in the position of the of the residues involved in the crosslink arising due to the multi-scale (coarse-grained) nature of the model representation is captured by the two sigma values corresponding to each residue involved in the crosslink.
Example for ihm_cross_link_restraint category
loop_
_ihm_cross_link_restraint.id
_ihm_cross_link_restraint.group_id
_ihm_cross_link_restraint.entity_id_1
_ihm_cross_link_restraint.asym_id_1
_ihm_cross_link_restraint.seq_id_1
_ihm_cross_link_restraint.comp_id_1
_ihm_cross_link_restraint.entity_id_2
_ihm_cross_link_restraint.asym_id_2
_ihm_cross_link_restraint.seq_id_2
_ihm_cross_link_restraint.comp_id_2
_ihm_cross_link_restraint.type
_ihm_cross_link_restraint.conditional_crosslink_flag
_ihm_cross_link_restraint.model_granularity
_ihm_cross_link_restraint.distance_threshold
_ihm_cross_link_restraint.psi
_ihm_cross_link_restraint.sigma_1
_ihm_cross_link_restraint.sigma_2
1 1 3 C 17 LYS 3 C 421 GLU EDC ALL by-residue 25.0 0.05 0.01 0.01
2 2 3 C 206 LYS 3 C 219 ASP EDC ALL by-residue 25.0 0.05 0.01 0.01
3 3 3 C 38 LYS 3 C 429 GLU EDC ALL by-residue 25.0 0.05 0.01 0.01
4 4 3 C 400 LYS 3 C 440 GLU EDC ALL by-residue 25.0 0.05 0.01 0.01
5 5 3 C 400 LYS 3 C 441 GLU EDC ALL by-residue 25.0 0.05 0.01 0.01
6 6 3 C 400 LYS 3 C 523 ASP EDC ALL by-residue 25.0 0.05 0.01 0.01
7 7 3 C 735 LYS 3 C 403 ASP EDC ALL by-residue 25.0 0.05 0.01 0.01
8 8 3 C 735 LYS 3 C 405 GLU EDC ALL by-residue 25.0 0.05 0.01 0.01
9 9 3 C 865 LYS 3 C 894 GLU EDC ALL by-residue 25.0 0.05 0.01 0.01
10 10 3 C 865 LYS 3 C 898 ASP EDC ALL by-residue 25.0 0.05 0.01 0.01
11 11 4 D 103 LYS 4 D 130 ASP EDC ALL by-residue 25.0 0.05 0.01 0.01
12 12 4 D 103 LYS 4 D 131 GLU EDC ALL by-residue 25.0 0.05 0.01 0.01
# ... abbreviated ...
Electron microscopy experiments provide images of macromolecules in multiple orientations, which are the reconstructed to obtain the three-dimensional density map. The structure of the macromolecule is then obtained from the density map using standard techniques.
2D images of class averages from electron microscopy experiments
An intermediate step in 3DEM reconstruction involves the averaging of individual 2DEM images of molecules in the same orientation. This improves the signal-to-noise ratio. Sometimes these class averaged 2DEM images can be used as restraints in the integrative modeling experiment. The 2DEM class averages can be archived in the EMPIAR repository, if their corresponding 3DEM maps are archived in EMDB repository. However, in the absence of 3DEM maps, these 2DEM class averages can be referenced via persistent digital object identifiers (DOI) as discussed in Section 1 above.
Figure 4. The class average of 2D micrographs obtained during an electron microscopy experiment can be used as a restraint in integrative modeling. 2D projections of sampled 3D models can be fit to the 2D class average image to provide a fit-to-image score. The 2DEM class average of the Nup-84 sub-complex from budding yeast is shown in the figure.
The ihm_2dem_class_average_restraint category captures the details of the 2dem class average image
used in the integrative modeling experiment. These details include image related parameters such as
pixel size, image resolution, number of raw micrographs used to obtain the image as well as other
implementation details such as whether the image is segmented to use only a part of it as restraint
in the modeling.
Example for ihm_2dem_class_average_restraint category
_ihm_2dem_class_average_restraint.id 1
_ihm_2dem_class_average_restraint.dataset_list_id 4
_ihm_2dem_class_average_restraint.number_raw_micrographs 200
_ihm_2dem_class_average_restraint.pixel_size_width 5.91
_ihm_2dem_class_average_restraint.pixel_size_height 5.91
_ihm_2dem_class_average_restraint.image_resolution 30.0
_ihm_2dem_class_average_restraint.image_segment_flag No
_ihm_2dem_class_average_restraint.number_of_projections 400
_ihm_2dem_class_average_restraint.struct_assembly_id 1
_ihm_2dem_class_average_restraint.details .
3D maps from electron microscopy experiments
The 3DEM maps are archived in the EMDB repository. All relevant information regarding the maps are defined in the EMDB entry. Here, we capture the information regarding how the maps are used as restraints in the integrative modeling. These include the fitting method used to fit the model to the 3DEM map, the citation id of the fitting method, if applicable and the cross-correlation coefficient corresponding to the fit.
Figure 5. 3D reconstructions obtained from electron microscopy experiments can be used as restraints in integrative modeling. Sampled models can be fit to the 3D maps to obtain a fit-to-density score that can be used to score the models.
Example for ihm_3dem_restraint category
_ihm_3dem_restraint.ordinal_id 1
_ihm_3dem_restraint.dataset_list_id 3
_ihm_3dem_restraint.model_id 1
_ihm_3dem_restraint.struct_assembly_id 1
_ihm_3dem_restraint.fitting_method 'Gaussian mixture models'
_ihm_3dem_restraint.number_of_gaussians 10
_ihm_3dem_restraint.map_segment_flag No
_ihm_3dem_restraint.cross_correlation_coefficient 0.8
_ihm_3dem_restraint.fitting_method_citation_id 5
_ihm_3dem_restraint.details .
Restraints from Small Angle Scattering (SAS) experiments
The data from SAS experiments are archived in the SASBDB repository.
All relevant information regarding the SAS experiment including the SAS profiles are captured in the
SASBDB entry. In the ihm_sas_restraint category, we capture the information regarding how the
SAS data is used as restraints in the integrative modeling. These include the fitting method used to
fit the model to the SAS profile, the chi-value corresponding to the fit and the radius of gyration
obtained from the SAS data.
Figure 6. An example of an experimentally derived SAS profile is shown (left, black dots) along with the theoretical SAS profile (left, red line) derived from a model obtained through integrative modeling (right).
Example for ihm_sas_restraint category
_ihm_sas_restraint.ordinal_id 1
_ihm_sas_restraint.dataset_list_id 6
_ihm_sas_restraint.model_id 1
_ihm_sas_restraint.struct_assembly_id 1
_ihm_sas_restraint.profile_segment_flag No
_ihm_sas_restraint.fitting_atom_type 'Heavy atoms'
_ihm_sas_restraint.fitting_method 'FoXS'
_ihm_sas_restraint.fitting_state 'Single'
_ihm_sas_restraint.radius_of_gyration 21.07
_ihm_sas_restraint.chi_value 4.69
_ihm_sas_restraint.details .
Integrative modeling of macromolecular assemblies may include starting structural models
of the individual components of the assembly obtained from the structural repositories
such as PDB and the model archive at
PMP. These starting structural models are
desribed in the ihm_starting_model_details, ihm_starting_comparative_models,
ihm_starting_model_seq_dif and ihm_starting_model_coord categories.
Figure 7. Starting structural models of individual components in a macromolecular assembly may be obtained through experimental methods such as X-ray crystallography and NMR spectroscopy or through computational methods such as abinitio or comparative modeling.
The ihm_starting_model_details category provides information on starting models used in the
integrative modeling.
Example for ihm_starting_model_details category
loop_
_ihm_starting_model_details.ordinal_id
_ihm_starting_model_details.entity_id
_ihm_starting_model_details.entity_description
_ihm_starting_model_details.asym_id
_ihm_starting_model_details.seq_id_begin
_ihm_starting_model_details.seq_id_end
_ihm_starting_model_details.starting_model_source
_ihm_starting_model_details.starting_model_auth_asym_id
_ihm_starting_model_details.starting_model_sequence_offset
_ihm_starting_model_details.starting_model_id
_ihm_starting_model_details.dataset_list_id
2 1 Nup84 A 7 436 'experimental model' C 0 Nup84-m1 5
3 1 Nup84 A 7 436 'experimental model' C 0 Nup84-m1 6
4 1 Nup84 A 429 488 'comparative model' A 0 Nup84-m1 7
6 1 Nup84 A 506 726 'comparative model' A 0 Nup84-m1 8
# ... abbreviated ...
The ihm_starting_comparative_models category captures additional details regarding the starting comparative
models used in the integrative modeling. The sequence alignments used to obtain the comparative models are
linked out to external files via the ihm_external_reference_info and ihm_external_files categories.
Example for ihm_starting_comparative_models category
loop_
_ihm_starting_comparative_models.starting_model_ordinal_id
_ihm_starting_comparative_models.starting_model_id
_ihm_starting_comparative_models.template_auth_asym_id
_ihm_starting_comparative_models.template_seq_begin
_ihm_starting_comparative_models.template_seq_end
_ihm_starting_comparative_models.template_sequence_identity
_ihm_starting_comparative_models.template_sequence_identity_denominator
_ihm_starting_comparative_models.template_dataset_list_id
_ihm_starting_comparative_models.alignment_file_id
4 Nup84-m1 G 482 551 10.0 1 9 7
6 Nup84-m1 A 667 924 18.0 1 10 6
# ... abbreviated ...
The ihm_starting_model_seq_dif category provides a mechanism for indicating and annotating point differences
between the sequence of the entity or biological unit in the current integrative model and the starting model
used in the integrative modeling referenced from a database (the Protein Data Bank or the Model Archive).
Example for ihm_starting_model_seq_dif category
loop_
_ihm_starting_model_seq_dif.ordinal_id
_ihm_starting_model_seq_dif.entity_id
_ihm_starting_model_seq_dif.asym_id
_ihm_starting_model_seq_dif.seq_id
_ihm_starting_model_seq_dif.comp_id
_ihm_starting_model_seq_dif.starting_model_ordinal_id
_ihm_starting_model_seq_dif.db_entity_id
_ihm_starting_model_seq_dif.db_asym_id
_ihm_starting_model_seq_dif.db_seq_id
_ihm_starting_model_seq_dif.db_comp_id
_ihm_starting_model_seq_dif.details
1 1 A 1 MET 2 1 A 1 MSE "Conversion of modified residue MSE to MET"
2 2 C 62 GLU 3 1 A 61 PHE "Point change of PHE to GLU"
# ...abbreviated...
The ihm_starting_model_coord records the details of the XYZ coordinates of the starting
models described in the ihm_starting_model_details category. This category is similar
to the atom_site category
in the PDBX/mmCIF dictionary. The coordinates of the starting models can also be linked out
to external files via the ihm_external_reference_info, ihm_external_files, ihm_dataset_list
and ihm_dataset_external_reference categories.
Example for ihm_starting_model_coord category
loop_
_ihm_starting_model_coord.starting_model_id
_ihm_starting_model_coord.group_PDB
_ihm_starting_model_coord.id
_ihm_starting_model_coord.type_symbol
_ihm_starting_model_coord.atom_id
_ihm_starting_model_coord.comp_id
_ihm_starting_model_coord.entity_id
_ihm_starting_model_coord.asym_id
_ihm_starting_model_coord.seq_id
_ihm_starting_model_coord.Cartn_x
_ihm_starting_model_coord.Cartn_y
_ihm_starting_model_coord.Cartn_z
_ihm_starting_model_coord.B_iso_or_equiv
_ihm_starting_model_coord.formal_charge
_ihm_starting_model_coord.ordinal_id
Nup84-m1 ATOM 1 N N TYR 1 A 7 -9.852 10.605 -6.336 91.82 . 1
Nup84-m1 ATOM 2 C CA TYR 1 A 7 -8.986 11.688 -5.817 91.82 . 2
Nup84-m1 ATOM 3 C CB TYR 1 A 7 -7.787 11.073 -5.072 91.82 . 3
Nup84-m1 ATOM 4 C CG TYR 1 A 7 -6.741 12.117 -4.873 91.82 . 4
Nup84-m1 ATOM 5 C CD1 TYR 1 A 7 -6.906 13.168 -3.998 91.82 . 5
Nup84-m1 ATOM 6 C CD2 TYR 1 A 7 -5.561 12.013 -5.571 91.82 . 6
Nup84-m1 ATOM 7 C CE1 TYR 1 A 7 -5.91 14.108 -3.843 91.82 . 7
Nup84-m1 ATOM 8 C CE2 TYR 1 A 7 -4.566 12.945 -5.421 91.82 . 8
Nup84-m1 ATOM 9 C CZ TYR 1 A 7 -4.738 13.997 -4.558 91.82 . 9
Nup84-m1 ATOM 10 O OH TYR 1 A 7 -3.708 14.949 -4.411 91.82 . 10
Nup84-m1 ATOM 11 C C TYR 1 A 7 -8.513 12.596 -6.909 91.82 . 11
# ... abbreviated ...
The integrative modeling can involve any combination of the following.
- Multi-scale representation involving atomistic and coarse-grained representations.
- Multi-state modeling involving multiple conformational states.
- Time-ordered ensemble representing multiple models related by time.
Figure 8. The integrative modeling hierarchy involving multi-scale, multi-state, time-ordered ensembles. The PDB archive currently holds only atomistic models using the PDBx/mmCIF dictionary. The new I/H methods dictionary systematically extends this to represent multi-scale, multi-state, time-ordered ensembles of I/H models.
The details of the multi-scale, multi-state, time-ordered modeling experiment is described in the categories discussed below.
Integrative models can include atomistic and coarse-grained representation as seen in the following figure.
Figure 9. Multi-scale modeling of macro-molecular assemblies include representations of the macrmolecules in atomistic and other coarse-grained representations such as beads or spheres corresponding to single residues or a set of residues as shown in the figure. The assembly can be a combination of various representations (both atomistic and non-atomistic), depending upon the kind of structural data that is available for the individual components in the assembly.
The representation of the multi-scale model is described in the ihm_model_representation
category. This category provides information on the various regions of the model that are
represented using different primitives (atomistic, spheres, gaussians), their mode
in the modeling experiment (rigid body or flexible regions) and the granularity of
the model (by-residue for atomistic representations and by-feature for spherical beads
and gaussian objects). The individual details of the atoms, spheres and Gaussians are
captured in other categories described in the following sections while this category
provides an overview of the model architecture. Current non-atomistic representations
include spherical beads and Gaussian objects. These can be further extended to include
other primitive representations as the integrative modeling field evolves.
Example for the ihm_model_representation category
loop_
_ihm_model_representation.ordinal_id
_ihm_model_representation.representation_id
_ihm_model_representation.segment_id
_ihm_model_representation.entity_id
_ihm_model_representation.entity_description
_ihm_model_representation.entity_asym_id
_ihm_model_representation.seq_id_begin
_ihm_model_representation.seq_id_end
_ihm_model_representation.model_object_primitive
_ihm_model_representation.starting_model_id
_ihm_model_representation.model_mode
_ihm_model_representation.model_granularity
_ihm_model_representation.model_object_count
1 1 1 1 Nup84 A 1 6 sphere Nup84-m1 flexible by-feature 1
2 1 2 1 Nup84 A 7 20 sphere Nup84-m1 rigid by-residue .
3 1 3 1 Nup84 A 21 26 sphere Nup84-m1 flexible by-feature 1
4 1 4 1 Nup84 A 27 80 sphere Nup84-m1 rigid by-residue .
5 1 5 1 Nup84 A 81 95 sphere Nup84-m1 flexible by-feature 1
6 1 6 1 Nup84 A 96 126 sphere Nup84-m1 rigid by-residue .
7 1 6 1 Nup84 A 96 126 sphere Nup84-m1 flexible by-feature 3
8 1 7 1 Nup84 A 127 135 sphere Nup84-m1 flexible by-feature 1
# ... abbreviated ...
Atomistic representations
Atomistic representations follow from the atom_site category in the PDBX/mmCIF dictionary. More details can be obtained from the atom_site description in the mmCIF dictionary and the mmCIF dictionary tutorials.
Example for the atom_site category obtained from pdbid 16pk.
loop_
_atom_site.group_PDB
_atom_site.id
_atom_site.type_symbol
_atom_site.label_atom_id
_atom_site.label_alt_id
_atom_site.label_comp_id
_atom_site.label_asym_id
_atom_site.label_entity_id
_atom_site.label_seq_id
_atom_site.pdbx_PDB_ins_code
_atom_site.Cartn_x
_atom_site.Cartn_y
_atom_site.Cartn_z
_atom_site.occupancy
_atom_site.B_iso_or_equiv
_atom_site.Cartn_x_esd
_atom_site.Cartn_y_esd
_atom_site.Cartn_z_esd
_atom_site.occupancy_esd
_atom_site.B_iso_or_equiv_esd
_atom_site.pdbx_formal_charge
_atom_site.auth_seq_id
_atom_site.auth_comp_id
_atom_site.auth_asym_id
_atom_site.auth_atom_id
_atom_site.pdbx_PDB_model_num
ATOM 1 N N . GLU A 1 1 ? -15.953 21.156 16.122 1.00 30.79 ? ? ? ? ? ? 5 GLU A N 1
ATOM 2 C CA . GLU A 1 1 ? -15.944 19.956 15.229 1.00 29.23 ? ? ? ? ? ? 5 GLU A CA 1
ATOM 3 C C . GLU A 1 1 ? -14.514 19.536 14.879 1.00 24.32 ? ? ? ? ? ? 5 GLU A C 1
ATOM 4 O O . GLU A 1 1 ? -14.204 18.353 14.856 1.00 24.00 ? ? ? ? ? ? 5 GLU A O 1
ATOM 5 C CB . GLU A 1 1 ? -16.724 20.230 13.940 1.00 34.27 ? ? ? ? ? ? 5 GLU A CB 1
ATOM 6 C CG . GLU A 1 1 ? -16.904 18.988 13.089 1.00 43.27 ? ? ? ? ? ? 5 GLU A CG 1
ATOM 7 C CD . GLU A 1 1 ? -17.018 19.289 11.606 1.00 49.49 ? ? ? ? ? ? 5 GLU A CD 1
ATOM 8 O OE1 . GLU A 1 1 ? -18.116 19.722 11.166 1.00 54.29 ? ? ? ? ? ? 5 GLU A OE1 1
ATOM 9 O OE2 . GLU A 1 1 ? -16.013 19.066 10.880 1.00 51.16 ? ? ? ? ? ? 5 GLU A OE2 1
ATOM 10 N N . LYS A 1 2 ? -13.659 20.505 14.561 1.00 20.45 ? ? ? ? ? ? 6 LYS A N 1
ATOM 11 C CA . LYS A 1 2 ? -12.271 20.198 14.247 1.00 17.38 ? ? ? ? ? ? 6 LYS A CA 1
ATOM 12 C C . LYS A 1 2 ? -11.523 20.071 15.575 1.00 15.75 ? ? ? ? ? ? 6 LYS A C 1
ATOM 13 O O . LYS A 1 2 ? -11.871 20.735 16.554 1.00 16.32 ? ? ? ? ? ? 6 LYS A O 1
ATOM 14 C CB . LYS A 1 2 ? -11.657 21.302 13.381 1.00 18.54 ? ? ? ? ? ? 6 LYS A CB 1
ATOM 15 C CG . LYS A 1 2 ? -12.506 21.632 12.152 1.00 20.47 ? ? ? ? ? ? 6 LYS A CG 1
ATOM 16 C CD . LYS A 1 2 ? -12.024 20.975 10.860 1.00 22.33 ? ? ? ? ? ? 6 LYS A CD 1
ATOM 17 C CE . LYS A 1 2 ? -11.851 19.494 10.969 1.00 21.49 ? ? ? ? ? ? 6 LYS A CE 1
ATOM 18 N NZ . LYS A 1 2 ? -11.636 18.855 9.646 1.00 17.34 ? ? ? ? ? ? 6 LYS A NZ 1
# ... abbreviated ...
Spherical representations
Spheres or beads representing single residues or multiple residues are captured using
the ihm_sphere_obj_site category.
Example for the ihm_sphere_obj_site category representing spherical beads comprising of single or multiple residues
loop_
_ihm_sphere_obj_site.ordinal_id
_ihm_sphere_obj_site.entity_id
_ihm_sphere_obj_site.seq_id_begin
_ihm_sphere_obj_site.seq_id_end
_ihm_sphere_obj_site.asym_id
_ihm_sphere_obj_site.Cartn_x
_ihm_sphere_obj_site.Cartn_y
_ihm_sphere_obj_site.Cartn_z
_ihm_sphere_obj_site.object_radius
_ihm_sphere_obj_site.rmsf
_ihm_sphere_obj_site.model_id
1 1 1 6 A 1.0 1.0 1.0 1.0 . 1
2 1 7 7 A 1.0 1.0 1.0 1.0 . 1
3 1 8 8 A 1.0 1.0 1.0 1.0 . 1
4 1 9 9 A 1.0 1.0 1.0 1.0 . 1
5 1 10 10 A 1.0 1.0 1.0 1.0 . 1
6 1 11 11 A 1.0 1.0 1.0 1.0 . 1
7 1 12 12 A 1.0 1.0 1.0 1.0 . 1
8 1 13 13 A 1.0 1.0 1.0 1.0 . 1
9 1 14 14 A 1.0 1.0 1.0 1.0 . 1
10 1 15 15 A 1.0 1.0 1.0 1.0 . 1
# ... abbreviated ...
Gaussian objects
Gaussian primitives representing single or multiple residues are captured using
the ihm_gaussian_obj_site category.
Figure 10. Modeling of ELF3 complex. An example of a Gaussian object used as a primitive representation in the multi-scale model.
Example for the ihm_gaussian_obj_site category representing 3D Gaussian objects in the model representation
loop_
_ihm_gaussian_obj_site.ordinal_id
_ihm_gaussian_obj_site.entity_id
_ihm_gaussian_obj_site.seq_id_begin
_ihm_gaussian_obj_site.seq_id_end
_ihm_gaussian_obj_site.asym_id
_ihm_gaussian_obj_site.mean_Cartn_x
_ihm_gaussian_obj_site.mean_Cartn_y
_ihm_gaussian_obj_site.mean_Cartn_z
_ihm_gaussian_obj_site.weight
_ihm_gaussian_obj_site.covariance_matrix[1][1]
_ihm_gaussian_obj_site.covariance_matrix[1][2]
_ihm_gaussian_obj_site.covariance_matrix[1][3]
_ihm_gaussian_obj_site.covariance_matrix[2][1]
_ihm_gaussian_obj_site.covariance_matrix[2][2]
_ihm_gaussian_obj_site.covariance_matrix[2][3]
_ihm_gaussian_obj_site.covariance_matrix[3][1]
_ihm_gaussian_obj_site.covariance_matrix[3][2]
_ihm_gaussian_obj_site.covariance_matrix[3][3]
_ihm_gaussian_obj_site.model_id
1 1 1 6 A 1.0 1.0 1.0 1.0 96.25 0.92 -30.50 0.92 83.82 16.37 -30.50 16.37 95.80 1
2 1 7 7 A 1.0 1.0 1.0 1.0 96.25 0.92 -30.50 0.92 83.82 16.37 -30.50 16.37 95.80 1
3 1 8 8 A 1.0 1.0 1.0 1.0 96.25 0.92 -30.50 0.92 83.82 16.37 -30.50 16.37 95.80 1
4 1 9 9 A 1.0 1.0 1.0 1.0 96.25 0.92 -30.50 0.92 83.82 16.37 -30.50 16.37 95.80 1
5 1 10 10 A 1.0 1.0 1.0 1.0 96.25 0.92 -30.50 0.92 83.82 16.37 -30.50 16.37 95.80 1
6 1 11 11 A 1.0 1.0 1.0 1.0 96.25 0.92 -30.50 0.92 83.82 16.37 -30.50 16.37 95.80 1
7 1 12 12 A 1.0 1.0 1.0 1.0 96.25 0.92 -30.50 0.92 83.82 16.37 -30.50 16.37 95.80 1
8 1 13 13 A 1.0 1.0 1.0 1.0 96.25 0.92 -30.50 0.92 83.82 16.37 -30.50 16.37 95.80 1
9 1 14 14 A 1.0 1.0 1.0 1.0 96.25 0.92 -30.50 0.92 83.82 16.37 -30.50 16.37 95.80 1
10 1 15 15 A 1.0 1.0 1.0 1.0 96.25 0.92 -30.50 0.92 83.82 16.37 -30.50 16.37 95.80 1
# ... abbreviated ...
Integrative modeling can involve molecular ensembles in different conformational states such as open and closed states, active and inactive states or ligand bound and unbound states.
Figure 11. Multi-state modeling may involve proteins in multiple states as shown in the figure. A. Calcium bound (left) and unbound (right) states of Calcium Pump. B. Inactive tensed (left) and active relaxed (right) states of Glycogen Phosphorylase. C. Open (left) and closed (right) states of Hexokinase. D. Active (left) and inactive (right) states of Lactate dehydrogenase. Pictures courtesy of PDB101.
Multiple states may exist simultaneously such that some or all of the collected experimental information
(e.g. crosslinks) reflect more than one state, and cannot be satisfied by a single model. Multi-state
modeling information is captured in the ihm_multi_state_modeling category, as shown in the following
examples. The multiple states can be grouped together to represent a collection of states that may evolve
with time.
Example 1: Conformational change
loop_
_ihm_multi_state_modeling.ordinal_id
_ihm_multi_state_modeling.state_id
_ihm_multi_state_modeling.state_group_id
_ihm_multi_state_modeling.population_fraction
_ihm_multi_state_modeling.state_type
_ihm_multi_state_modeling.state_name
_ihm_multi_state_modeling.model_group_id
_ihm_multi_state_modeling.experiment_type
_ihm_multi_state_modeling.details
1 1 1 0.5 'conformational change' 'open' 1 'Fraction of bulk' 'open state ensemble 1'
2 2 1 0.5 'conformational change' 'closed' 2 'Fraction of bulk' 'closed state ensemble 2'
# ... abbreviated ...
Example 2: Ligand binding
loop_
_ihm_multi_state_modeling.ordinal_id
_ihm_multi_state_modeling.state_id
_ihm_multi_state_modeling.state_group_id
_ihm_multi_state_modeling.population_fraction
_ihm_multi_state_modeling.state_type
_ihm_multi_state_modeling.state_name
_ihm_multi_state_modeling.model_group_id
_ihm_multi_state_modeling.experiment_type
_ihm_multi_state_modeling.details
1 1 1 0.25 'ligand binding' 'unbound' 1 'Fraction of bulk' 'unbound'
2 2 1 0.75 'ligand binding' 'bound' 2 'Fraction of bulk' 'bound to heme'
# ... abbreviated ...
Example 3: Complex formation
loop_
_ihm_multi_state_modeling.ordinal_id
_ihm_multi_state_modeling.state_id
_ihm_multi_state_modeling.state_group_id
_ihm_multi_state_modeling.population_fraction
_ihm_multi_state_modeling.state_type
_ihm_multi_state_modeling.state_name
_ihm_multi_state_modeling.model_group_id
_ihm_multi_state_modeling.experiment_type
_ihm_multi_state_modeling.details
1 1 1 . 'complex formation' 'unbound' 1 'Fraction of bulk' 'unbound molecule 1'
2 2 1 . 'complex formation' 'unbound' 2 'Fraction of bulk' 'unbound molecule 2'
3 3 1 . 'complex formation' 'bound' 3 'Fraction of bulk' 'bound molecules 1 and 2'
# ... abbreviated ...
Integrative modeling can involve time-ordered ensembles such as macromolecules in reaction pathways, metabolic networks, trajectories or any other time-ordered relationships. These can be related in a linear, branched or cyclic manner.
Figure 12. Time-ordered ensembles may involve macromolecular complexes involved in a reaction pathway or metabolic network as shown in the figure. A. Enzymes involved in bacterial fatty acid biosynthesis pathway. B. Heterotrimeric G proteins involved in signal transduction pathway. Pictures are courtesy of PDB101.
The details of time-ordered ensembles are captured in the ihm_time_ordered_ensemble
category as shown in the examples given below. The time-ordered relationships are stored
as a simple directed graph of the models, ordered in time and connected by a set of edges.
The start and end points of the time-ordered process is listed along with the model groups
and/or state groups that represent the start and end points. The time unit can be included,
if relevant.
Figure 13. Time-ordered ensembled are represented as directed graphs using ensembles or states of models as nodes and the time-ordered relationship between them as edges.
Example 1: Linear process
loop_
_ihm_time_ordered_ensemble.ordinal_id
_ihm_time_ordered_ensemble.time_step_id
_ihm_time_ordered_ensemble.time_begin
_ihm_time_ordered_ensemble.time_end
_ihm_time_ordered_ensemble.time_unit
_ihm_time_ordered_ensemble.model_group_id_begin
_ihm_time_ordered_ensemble.model_group_id_end
_ihm_time_ordered_ensemble.state_group_id_begin
_ihm_time_ordered_ensemble.state_group_id_end
1 1 0 1 nanosecond . . 1 2
2 2 1 2 nanosecond . . 2 3
3 3 2 3 nanosecond . . 3 4
4 4 3 4 nanosecond . . 4 5
# ... abbreviated ...
Example 2: Cyclic process
loop_
_ihm_time_ordered_ensemble.ordinal_id
_ihm_time_ordered_ensemble.time_step_id
_ihm_time_ordered_ensemble.time_begin
_ihm_time_ordered_ensemble.time_end
_ihm_time_ordered_ensemble.time_unit
_ihm_time_ordered_ensemble.model_group_id_begin
_ihm_time_ordered_ensemble.model_group_id_end
_ihm_time_ordered_ensemble.state_group_id_begin
_ihm_time_ordered_ensemble.state_group_id_end
1 1 0 1 nanosecond 1 2 . .
2 2 1 2 nanosecond 2 3 . .
3 3 2 3 nanosecond 3 4 . .
4 4 3 4 nanosecond 4 1 . .
# ... abbreviated ...
Example 3: Branched process
loop_
_ihm_time_ordered_ensemble.ordinal_id
_ihm_time_ordered_ensemble.time_step_id
_ihm_time_ordered_ensemble.time_begin
_ihm_time_ordered_ensemble.time_end
_ihm_time_ordered_ensemble.time_unit
_ihm_time_ordered_ensemble.model_group_id_begin
_ihm_time_ordered_ensemble.model_group_id_end
_ihm_time_ordered_ensemble.state_group_id_begin
_ihm_time_ordered_ensemble.state_group_id_end
1 1 0 1 picosecond 1 2 . .
2 2 1 2 picosecond 2 3 . .
3 2 1 2 picosecond 2 4 . .
4 3 2 3 picosecond 3 5 . .
5 3 2 3 picosecond 4 6 . .
# ... abbreviated ...
Preliminary information on the integrative modeling workflow and the post-modeling
process are captured in the ihm_modeling_protocol and the ihm_modeling_post_process
categories. The ihm_modeling_protocol category provides references to the input
dataset described in the ihm_dataset_list category and to the structural assembly of
macromolecules described in the ihm_struct_assembly category. It also includes flags
to indicate whether the integrative modeling experiment involves multi-scale, multi-state
and time-ordered ensembles. The various steps involved in the modeling and post-process,
such as sampling, scoring, filtering and clustering can be described in these categories.
The scripts and other workflow related files stored externally are included in the
ihm_modeling_workflow_files category, which links to the ihm_external_reference_info
category via the ihm_external_files category.
Figure 14. An example of the modeling workflow is shown. The modeling task may involve various sampling, scoring and post-modeling analyses steps that are briefly captured in the IHM dictionary extension.
Example 1 for the ihm_modeling_protocol category
_ihm_modeling_protocol.ordinal_id
_ihm_modeling_protocol.protocol_id
_ihm_modeling_protocol.step_id
_ihm_modeling_protocol.struct_assembly_id
_ihm_modeling_protocol.dataset_group_id
_ihm_modeling_protocol.struct_assembly_description
_ihm_modeling_protocol.protocol_name
_ihm_modeling_protocol.step_name
_ihm_modeling_protocol.step_method
_ihm_modeling_protocol.num_models_begin
_ihm_modeling_protocol.num_models_end
_ihm_modeling_protocol.multi_scale_flag
_ihm_modeling_protocol.multi_state_flag
_ihm_modeling_protocol.time_ordered_flag
1 1 1 1 1 'Nup84 sub-complex' 'Multi-scale modeling' 'Sampling' 'Replica exchange monte carlo' 0 7500 Y N N
2 1 2 1 2 'Nup84 sub-complex' 'Multi-scale modeling' 'Sampling' 'Replica exchange monte carlo' 7500 15000 Y N N
# ... abbreviated ...
Example 2 for the ihm_modeling_protocol category
_ihm_modeling_protocol.ordinal_id
_ihm_modeling_protocol.protocol_id
_ihm_modeling_protocol.step_id
_ihm_modeling_protocol.struct_assembly_id
_ihm_modeling_protocol.dataset_group_id
_ihm_modeling_protocol.struct_assembly_description
_ihm_modeling_protocol.protocol_name
_ihm_modeling_protocol.step_name
_ihm_modeling_protocol.step_method
_ihm_modeling_protocol.num_models_begin
_ihm_modeling_protocol.num_models_end
_ihm_modeling_protocol.multi_scale_flag
_ihm_modeling_protocol.multi_state_flag
_ihm_modeling_protocol.time_ordered_flag
1 1 1 1 1 'PhoQ' 'Multi-state modeling' 'Sampling' 'Monte carlo' 0 5000 N Y N
2 1 1 2 1 'PhoQ' 'Multi-state modeling' 'Sampling' 'Monte carlo' 0 5000 N Y N
# ... abbreviated ...
Example for the ihm_modeling_post_process category
loop_
_ihm_modeling_post_process.id
_ihm_modeling_post_process.protocol_id
_ihm_modeling_post_process.analysis_id
_ihm_modeling_post_process.step_id
_ihm_modeling_post_process.type
_ihm_modeling_post_process.feature
_ihm_modeling_post_process.num_models_begin
_ihm_modeling_post_process.num_models_end
1 1 1 1 'filter' 'energy/score' 15000 6520
2 1 1 2 'cluster' 'dRMSD' 6520 6520
3 1 2 1 'filter' 'energy/score' 15000 6520
4 1 2 2 'filter' 'composition' 6520 6520
5 1 2 3 'cluster' 'dRMSD' 6520 6520
# ... abbreviated ...
The ihm_model_list category provides an inventory list of the models deposited,
along with references to the modeling protocol that produced the models and the
structure assembly/sub-assembly corresponding to the model. The models can be grouped
together according to the discretion of the depositor. The grouping can refer to any
combination of the following and can have an appropriate user-friendly name.
- Clusters or ensembles resulting from the integrative modeling experiment
- Different states in a multi-state modeling experiment
- Models from a particular time step in a time-ordered ensemble
Some examples of the ihm_model_list category are given below.
Example 1:
loop_
_ihm_model_list.ordinal_id
_ihm_model_list.model_id
_ihm_model_list.model_group_id
_ihm_model_list.model_name
_ihm_model_list.model_group_name
_ihm_model_list.assembly_id
_ihm_model_list.protocol_id
1 1 1 'Best scoring model' 'ensemble1' 1 1
2 2 1 '2nd Best scoring model' 'ensemble1' 1 1
3 3 1 '3rd Best scoring model' 'ensemble1' 1 1
4 4 1 '4th Best scoring model' 'ensemble1' 1 1
5 5 1 '5th Best scoring model' 'ensemble1' 1 1
6 6 2 'Best scoring model' 'ensemble2' 1 1
7 7 2 '2nd Best scoring model' 'ensemble2' 1 1
8 8 2 '3rd Best scoring model' 'ensemble2' 1 1
9 9 2 '4th Best scoring model' 'ensemble2' 1 1
10 10 2 '5th Best scoring model' 'ensemble2' 1 1
# ... abbreviated ...
Example 2:
loop_
_ihm_model_list.ordinal_id
_ihm_model_list.model_id
_ihm_model_list.model_group_id
_ihm_model_list.model_group_name
_ihm_model_list.assembly_id
_ihm_model_list.protocol_id
1 1 1 'open state' 1 1
2 2 1 'open state' 1 1
3 3 1 'open state' 1 1
4 4 1 'open state' 1 1
5 5 1 'open state' 1 1
6 6 2 'closed state' 1 1
7 7 2 'closed state' 1 1
8 8 2 'closed state' 1 1
9 9 2 'closed state' 1 1
10 10 2 'closed state' 1 1
# ... abbreviated ...
Example 3:
loop_
_ihm_model_list.ordinal_id
_ihm_model_list.model_id
_ihm_model_list.model_group_id
_ihm_model_list.model_group_name
_ihm_model_list.assembly_id
_ihm_model_list.protocol_id
1 1 1 'Bound state ensemble 1' 1 1
2 2 1 'Bound state ensemble 1' 1 1
3 3 1 'Bound state ensemble 1' 1 1
4 4 1 'Bound state ensemble 1' 1 1
5 5 1 'Bound state ensemble 1' 1 1
6 6 2 'Unbound state ensemble 2' 2 1
7 7 2 'Unbound state ensemble 2' 2 1
8 8 2 'Unbound state ensemble 2' 2 1
9 9 2 'Unbound state ensemble 2' 2 1
10 10 2 'Unbound state ensemble 2' 2 1
# ... abbreviated ...
ihm_struct_assembly category describes the macromolecular assemblies and sub-assemblies that are
modeled by the current integrative modeling task. Sometimes, the integrative modeling task may lead
to structures of partial assemblies rather than the entire assembly that is used in the original experiments.
The structural assemblies and sub-assemblies are described in the ihm_struct_assembly category.
These sub-assemblies can consist of any combination of sequence ranges within a set of macromolecular
instances.
Figure 15. The macromolecular assembly comprises of various individual components that are
defined in the ihm_struct_assembly category. The integrative modeling task may involve
the entire assembly or a sub-assembly and these can be defined accordingly in the
ihm_struct_assembly category.
Example for the ihm_struct_assembly category.
loop_
_ihm_struct_assembly.ordinal_id
_ihm_struct_assembly.assembly_id
_ihm_struct_assembly.entity_description
_ihm_struct_assembly.entity_id
_ihm_struct_assembly.asym_id
_ihm_struct_assembly.seq_id_begin
_ihm_struct_assembly.seq_id_end
1 1 Nup84 1 A 1 726
2 1 Nup85 2 B 1 744
3 1 Nup120 3 C 1 1037
4 1 Nup133 4 D 1 1157
5 1 Nup145 5 E 1 712
6 1 Seh1 6 F 1 349
7 1 Sec13 7 G 1 297
8 2 Nup84 1 A 1 726
9 2 Nup85 2 B 1 744
10 2 Nup145 5 E 1 712
11 2 Seh1 6 F 1 349
12 2 Sec13 7 G 1 297
# ... abbreviated ...
ihm_ensemble_info category provides information on the ensembles of macromolecular
assemblies obtained from integrative modeling. This includes any clustering information
if applicable, number of models in the ensemble and information on the number of
representative models from the ensemble deposited in the archive. Some examples of
the ihm_ensemble_info category are shown below.
Figure 16. The integrative modeling experiment may lead to an ensemble of structural models
as shown in the figure. While only a subset of individual models may be deposited to the archive,
the ensembles themselves are described in the ihm_ensemble_info category.
Example 1:
loop_
_ihm_ensemble_info.ensemble_id
_ihm_ensemble_info.ensemble_name
_ihm_ensemble_info.post_process_id
_ihm_ensemble_info.model_group_id
_ihm_ensemble_info.ensemble_clustering_method
_ihm_ensemble_info.ensemble_clustering_feature
_ihm_ensemble_info.num_ensemble_models
_ihm_ensemble_info.num_ensemble_models_deposited
_ihm_ensemble_info.ensemble_precision_value
_ihm_ensemble_info.ensemble_file_id
1 'ensemble1' 2 1 Hierarchical dRMSD 2885 5 17.3 20
2 'ensemble2' 2 2 Hierarchical dRMSD 1401 5 19.2 21
3 'ensemble3' 5 . Hierarchical dRMSD 577 0 14.8 22
4 'ensemble4' 5 . Hierarchical dRMSD 404 0 10.8 23
# ... abbreviated ...
Example 2:
loop_
_ihm_ensemble_info.ensemble_id
_ihm_ensemble_info.ensemble_name
_ihm_ensemble_info.post_process_id
_ihm_ensemble_info.model_group_id
_ihm_ensemble_info.ensemble_clustering_method
_ihm_ensemble_info.ensemble_clustering_feature
_ihm_ensemble_info.num_ensemble_models
_ihm_ensemble_info.num_ensemble_models_deposited
_ihm_ensemble_info.ensemble_precision_value
_ihm_ensemble_info.ensemble_file_id
1 'open state' . 1 Hierarchical RMSD 2885 5 12.3 20
2 'closed state' . 2 Hierarchical RMSD 577 5 11.8 21
# ... abbreviated ...
The models obtained from integrative/hybrid methods can be validated against the set of experimentally-derived input restraints used in the modeling. Some preliminary validation parameters are included in the current dictionary and are discussed below.
Validation of crosslinks
The ihm_cross_link_result provides information on whether the crosslinking
restraints used in the modeling and defined in the ihm_cross_link_restraint
category are satisfied by the ensembles obtained from the modeling and
described in the ihm_ensemble_info category.
Example for the ihm_cross_link_result category
loop_
_ihm_cross_link_result.ordinal_id
_ihm_cross_link_result.restraint_id
_ihm_cross_link_result.ensemble_id
_ihm_cross_link_result.num_models
_ihm_cross_link_result.distance_threshold
_ihm_cross_link_result.median_distance
_ihm_cross_link_result.details
1 1 1 2885 25.0 9.6 .
2 2 1 2885 25.0 14.7 .
3 3 1 2885 25.0 16.4 .
4 4 1 2885 25.0 15.8 .
5 5 1 2885 25.0 18.9 .
6 6 1 2885 25.0 11.6 .
# ... abbreviated ...
The ihm_cross_link_result_parameters category holds the details regarding the
crosslinking parameters (psi, sigma_1 and sigma_2) for every cross link restraint
defined in the ihm_cross_link_restraint category and the corresponding integrative
model obtained at the end of the modeling experiment.
Example for the ihm_cross_link_result_parameters category
loop_
_ihm_cross_link_result_parameters.ordinal_id
_ihm_cross_link_result_parameters.restraint_id
_ihm_cross_link_result_parameters.model_id
_ihm_cross_link_result_parameters.psi
_ihm_cross_link_result_parameters.sigma_1
_ihm_cross_link_result_parameters.sigma_2
1 1 1 0.05 0.01 0.01
2 2 1 0.05 0.01 0.01
3 3 1 0.05 0.01 0.01
4 4 2 0.05 0.01 0.01
5 5 3 0.05 0.01 0.01
6 6 4 0.05 0.01 0.01
# ... abbreviated ...
Fitting of 2DEM data
The ihm_2dem_class_average_fitting category defines the transformation matrix
that can be used to fit a deposited model to the 2DEM class average image used
as an input restraint in the integrative modeling. It also provides a fit-to-image
correlation co-efficient that can be used to estimate the quality of the fit.
A matrix is provided for every 2DEM class average used as input and corresponds
to fitting a single deposited model.
Figure 17. The 2D projections of the sampled models are fit to the 2DEM class average to obtain a fit-to-image score.
Example for the ihm_2dem_class_average_fitting category
_ihm_2dem_class_average_fitting.ordinal_id 1
_ihm_2dem_class_average_fitting.restraint_id 1
_ihm_2dem_class_average_fitting.model_id 1
_ihm_2dem_class_average_fitting.cross_correlation_coefficient 0.9
_ihm_2dem_class_average_fitting.rot_matrix[1][1] 1.0
_ihm_2dem_class_average_fitting.rot_matrix[2][1] 0.0
_ihm_2dem_class_average_fitting.rot_matrix[3][1] 0.0
_ihm_2dem_class_average_fitting.rot_matrix[1][2] 0.0
_ihm_2dem_class_average_fitting.rot_matrix[2][2] 1.0
_ihm_2dem_class_average_fitting.rot_matrix[3][2] 0.0
_ihm_2dem_class_average_fitting.rot_matrix[1][3] 0.0
_ihm_2dem_class_average_fitting.rot_matrix[2][3] 0.0
_ihm_2dem_class_average_fitting.rot_matrix[3][3] 1.0
_ihm_2dem_class_average_fitting.tr_vector[1] 0.0
_ihm_2dem_class_average_fitting.tr_vector[2] 0.0
_ihm_2dem_class_average_fitting.tr_vector[3] 0.0
Localization densities of protein domains
The ihm_gaussian_object_ensemble category captures the information on
the localization densities of protein domains in the assembly, obtained
from the ensembles of sampled conformations reported in the ihm_ensemble_info
category. The Gaussian objects are alternate means of representing three-dimensional
densities of ensembles. This category is similar to the ihm_gaussian_obj_site
category used in the multi-scale representation of the integrative model.
Figure 18. The ensembles described in the ihm_ensemble_info category can provide
the localization densities of protein domains sampled during integrative modeling.
These densities can be represented as 3D Gaussians and are captured in the
ihm_gaussian_object_ensemble category.
Example for the ihm_gaussian_object_ensemble category.
loop_
_ihm_gaussian_obj_ensemble.ordinal_id
_ihm_gaussian_obj_ensemble.entity_id
_ihm_gaussian_obj_ensemble.seq_id_begin
_ihm_gaussian_obj_ensemble.seq_id_end
_ihm_gaussian_obj_ensemble.asym_id
_ihm_gaussian_obj_ensemble.mean_Cartn_x
_ihm_gaussian_obj_ensemble.mean_Cartn_y
_ihm_gaussian_obj_ensemble.mean_Cartn_z
_ihm_gaussian_obj_ensemble.weight
_ihm_gaussian_obj_ensemble.covariance_matrix[1][1]
_ihm_gaussian_obj_ensemble.covariance_matrix[1][2]
_ihm_gaussian_obj_ensemble.covariance_matrix[1][3]
_ihm_gaussian_obj_ensemble.covariance_matrix[2][1]
_ihm_gaussian_obj_ensemble.covariance_matrix[2][2]
_ihm_gaussian_obj_ensemble.covariance_matrix[2][3]
_ihm_gaussian_obj_ensemble.covariance_matrix[3][1]
_ihm_gaussian_obj_ensemble.covariance_matrix[3][2]
_ihm_gaussian_obj_ensemble.covariance_matrix[3][3]
_ihm_gaussian_obj_ensemble.ensemble_id
1 1 1 6 A 1.0 1.0 1.0 1.0 96.25 0.92 -30.50 0.92 83.82 16.37 -30.50 16.37 95.80 1
2 1 7 7 A 1.0 1.0 1.0 1.0 96.25 0.92 -30.50 0.92 83.82 16.37 -30.50 16.37 95.80 1
3 1 8 8 A 1.0 1.0 1.0 1.0 96.25 0.92 -30.50 0.92 83.82 16.37 -30.50 16.37 95.80 1
4 1 9 9 A 1.0 1.0 1.0 1.0 96.25 0.92 -30.50 0.92 83.82 16.37 -30.50 16.37 95.80 1
5 1 10 10 A 1.0 1.0 1.0 1.0 96.25 0.92 -30.50 0.92 83.82 16.37 -30.50 16.37 95.80 1
6 1 11 11 A 1.0 1.0 1.0 1.0 96.25 0.92 -30.50 0.92 83.82 16.37 -30.50 16.37 95.80 1
7 1 12 12 A 1.0 1.0 1.0 1.0 96.25 0.92 -30.50 0.92 83.82 16.37 -30.50 16.37 95.80 1
8 1 13 13 A 1.0 1.0 1.0 1.0 96.25 0.92 -30.50 0.92 83.82 16.37 -30.50 16.37 95.80 1
9 1 14 14 A 1.0 1.0 1.0 1.0 96.25 0.92 -30.50 0.92 83.82 16.37 -30.50 16.37 95.80 1
10 1 15 15 A 1.0 1.0 1.0 1.0 96.25 0.92 -30.50 0.92 83.82 16.37 -30.50 16.37 95.80 1
# ... abbreviated ...
Alternately, localization densities can be stored in an external file in a standard
format (such as CCP4 or MRC). Details of such externally stored localization density
files are provided in the ihm_localization_density_files category, which is linked
to the ihm_external_reference_info category via ihm_external_files category.
Example for the ihm_localization_density_files category
loop_
_ihm_localization_density_files.id
_ihm_localization_density_files.file_id
_ihm_localization_density_files.ensemble_id
_ihm_localization_density_files.entity_id
_ihm_localization_density_files.asym_id
_ihm_localization_density_files.seq_id_begin
_ihm_localization_density_files.seq_id_end
1 12 1 1 A 1 726
2 13 1 2 B 1 744
3 14 1 3 C 1 1037
# ...abbreviated...
Since the integrative/hybrid methods dictionary is designed to be an extension to the existing PDBX/mmCIF dictionary, many generic categories from the PDBX/mmCIF dictionary are highly relevant to models obtained from integrative/hybrid methods. Some of the most common categories are listed below and linked to their original decriptions on the mmCIF dictionary site.
The entry category
includes the _entry.id, which corresponds to the PDB id of the entry.
The entity category
includes details of all the molecular entitites in the modeled structure.
The entity_poly category
describes details of all the polymer entitites in the modeled structure.
The entity_poly_seq
category provides details of the sequence of monomers in the polymer entitites described in entity_poly.
The struct_asym category
records the details of the structural elements in the asymmetric unit and accounts for the strand id / chain id
corresponding to multiple instances of entities in the structural assembly.
The citation
category captures the details regarding the primary and other citations of the integrative modeling experiment.
The citation_author
category provides information of the authors in the citations listed in the citation category.
The software category
provides details of the various software used in the modeling, including their versions and other relevant information.
As mentioned before, the atom_site
category captures the information on the atomic XYZ coordinates.
| Category name | Category details |
|---|---|
ihm_dataset_list |
Inventory list of input data |
ihm_related_datasets |
Details of datasets derived from other datasets |
ihm_dataset_related_db_reference |
References to data from other repositories |
ihm_dataset_external_reference |
Details of data from external sources |
ihm_external_reference_info |
References to external data sources |
ihm_external_files |
Details of external files that store information |
ihm_cross_link_list |
List of crosslinks from experiments |
ihm_cross_link_restraint |
Crosslinking restraints implemented |
ihm_2dem_class_average_restraint |
Restraints from 2DEM class average image |
ihm_3dem_restraint |
Restraint from 3DEM maps |
ihm_sas_restraint |
Restraint from SAS data |
ihm_starting_model_details |
Details of starting structural models |
ihm_starting_comparative_models |
Details of starting comparative models |
ihm_starting_model_seq_dif |
Details of point differences in the sequences |
ihm_starting_model_coords |
XYZ coordinates of starting models |
ihm_model_representation |
Representation of multi-scale models |
ihm_sphere_obj_site |
Description of spherical objects in the model |
ihm_gaussian_obj_site |
Description of Gaussian objects in the model |
ihm_multi_state_modeling |
Description of multiple conformational states |
ihm_time_ordered_ensembles |
Description time-related models |
ihm_modeling_protocol |
Description of modeling workflow |
ihm_modeling_post_process |
Description of post-modeling processing |
ihm_model_list |
Inventory list of deposited models |
ihm_struct_assembly |
Description of assemblies and sub-assemblies |
ihm_ensemble_info |
Description of ensembles or clusters or models |
ihm_cross_link_result |
Description of satisfied or violated crosslinks |
ihm_cross_link_result_parameters |
Results of the crosslinking parameters |
ihm_2dem_class_average_fitting |
Fitting of model to 2DEM image |
ihm_gaussian_obj_ensemble |
Localization densities of protein domains |
ihm_localization_density_files |
Localization densities stored in external files |