CAPRI Evaluation module

Submodules

Module contents

Calculate CAPRI metrics for the input models.

By default the following metrics are calculated:

  • FNAT (fraction of native contacts), namely the fraction of

    intermolecular contacts in the docked complex that are also present in the reference complex.

  • IRMSD (interface root mean square deviation), namely the RMSD

    of the interface of the docked complex with respect to the reference complex.

  • LRMSD (ligand root mean square deviation), namely the RMSD of the

    ligand of the docked complex with respect to the reference complex upon superposition of the receptor.

  • DOCKQ, a measure of the quality of the docked model obtained

    by combining FNAT, IRMSD and LRMSD (see Basu and Wallner 2016, 11 (8), e0161879).

  • ILRMSD (interface ligand root mean square deviation), the RMSD of the

    ligand of the docked complex with respect to the reference complex upon superposition of the interface of the receptor.

  • GLOBAL_RMSD, the full RMSD between the reference and the model.

The following files are generated:

  • capri_ss.tsv: a table with the CAPRI metrics for each model.

  • capri_clt.tsv: a table with the CAPRI metrics for each cluster of models (if clustering information is available).

class haddock.modules.analysis.caprieval.HaddockModule(order: int, path: Path, init_params: str | Path = PosixPath('/opt/hostedtoolcache/Python/3.10.15/x64/lib/python3.10/site-packages/haddock/modules/analysis/caprieval/defaults.yaml'))[source]

Bases: BaseHaddockModule

HADDOCK3 module to calculate the CAPRI metrics.

classmethod confirm_installation() → None[source]

Confirm if contact executable is compiled.

static is_nested(models: list[PDBFile | list[PDBFile]]) → bool[source]
name: str = 'caprieval'

Default Parameters

Easy

alignment_method

default: ‘sequence’
type: string
title: Alignment method used to match the numbering.
choices: [‘structure’, ‘sequence’]
short description: Which alignment method should be used to match the numbering.
long description: Alignment method used to match the numbering. Sequence alignment is haddock3-friendly but might not produce best results for structures that are too different, for that use
group: analysis
explevel: easy

allatoms

default: False
type: boolean
title: Atoms to be considered during the analysis.
short description: Atoms to be considered during the analysis.
long description: Atoms to be considered during the analysis. If false (default), only backbone atoms will be considered, otherwise all the heavy-atoms.
group: analysis
explevel: easy

clt_threshold

default: 4
type: integer
title: Threshold (n) used for the average evaluation in the cluster-based output
min: 1
max: 1000
short description: Threshold (n) used for the average evaluation in the cluster-based output.
long description: Threshold (n) used for the average evaluation in the cluster-based output.
group: analysis
explevel: easy

dockq

default: True
type: boolean
title: Calculate DockQ
short description: Evaluate DockQ.
long description: Evaluate DockQ. A single continuous quality measure for protein docked models based on the CAPRI criteria, a combination of i-RMSD, l-RMSD and FNAT value. For details check the DockQ publication DOI:10.1371/journal.pone.0161879
group: analysis
explevel: easy

fnat

default: True
type: boolean
title: Calculate FNAT
short description: Performs FNAT calculations.
long description: Calculates the Fraction of Native Contacts. This evaluates how many contacts observed in the reference are present in the model. The distance cutoff used to define intermolecular contacts is defined by the fnat_cutoff parameter.
group: analysis
explevel: easy

fnat_cutoff

default: 5.0
type: float
title: Distance cutoff (Å) used to define interface contacts.
min: 3.0
max: 20.0
short description: Distance cutoff (Å) used to define interface contacts between two interacting molecules.
long description: Distance cutoff (Å) used to define interface contacts between two interacting molecules.
group: analysis
explevel: easy

global_rmsd

default: True
type: boolean
title: Calculate global RMSD
short description: Calculate full structure RMSD.
long description: Compute full structure RMSD between reference structure and the model after superimposition. Note that a sequence alignment is performed to obtain the mapping between the reference and the model. If the one of the structure contain a ‘small-molecule’, we assume that the compared structure atoms are in the same order. With default parameters allatoms = false, only backbone RMSD will be computed.
group: analysis
explevel: easy

ilrmsd

default: True
type: boolean
title: Calculate I-L-RMSD
short description: Performs Interface Ligand RMSD calculations.
long description: Calculates the Interface Ligand Root Mean Square Deviation. This calculation is done by superimposing the model and the target on the interface backbone atoms of the receptor (defined by the receptor_chain parameter) and calculating the RMSD on all heavy atoms of the ligand (defined by the ligand_chains parameter).
group: analysis
explevel: easy

irmsd

default: True
type: boolean
title: Calculate I-RMSD
short description: Performs Interface RMSD calculations.
long description: Calculates the Interface Root Mean Square Deviation. This calculate the RMSD of interface backbone atoms after fitting on the interface backbone atoms. The interface is idenfied based on intermolecular contacts shorter than the defined irmsd_cutoff.
group: analysis
explevel: easy

irmsd_cutoff

default: 10.0
type: float
title: Distance cutoff (Å) used to define interface residues.
min: 3.0
max: 20.0
short description: Distance cutoff (Å) used to define interface residues based on intermolecular contacts.
long description: Distance cutoff (Å) used to define interface residues based on intermolecular contacts.
group: analysis
explevel: easy

ligand_chains

default: []
type: list
title: Ligand ChainIDs
short description: Ligand ChainIDs to be considered for the RMSD calculations.
long description: Ligand ChainIDs to be considered for the RMSD calculations. This determines which chains will be treated as the ligand for the L-RMSD and for I-L-RMSD.
group: analysis
explevel: easy

lovoalign_exec

default: ‘’
type: string
title: Location (path) of the LovoAlign executable
short description: Location (path) of the LovoAlign executable.
long description: Location (path) of the LovoAlign executable.
group: analysis
explevel: easy

lrmsd

default: True
type: boolean
title: Calculate L-RMSD
short description: Performs Ligand RMSD calculations.
long description: Calculates the Ligand Root Mean Square Deviation. This calculation is done by superimposing the receptors (defined by the receptor_chain parameter) and then evaluating the RMSD on the ligands (defined by the ligand_chains parameter). Superposition and RMSD calculations are done on backbone heavy atoms.
group: analysis
explevel: easy

receptor_chain

default: ‘A’
type: string
title: Receptor ChainID
short description: Receptor ChainID to be considered for the RMSD calculations.
long description: Receptor ChainID to be considered for the RMSD calculations. This determines which chain will be treated as the receptor for the L-RMSD and for I-L-RMSD.
group: analysis
explevel: easy

reference_fname

default: ‘’
type: file
title: Reference structure
short description: Structure to be used when calculating the CAPRI metrics.
long description: Reference tructure to be used when calculating the CAPRI metrics. If none is defined then the lowest scoring model is selected by default.
group: analysis
explevel: easy

sort_ascending

default: True
type: boolean
title: Sort in ascending order
short description: Sort in ascending order.
long description: Sort in ascending order.
group: analysis
explevel: easy

sortby

default: ‘score’
type: string
title: Sort the structures/clusters in the output file by this value
choices: [‘score’, ‘irmsd’, ‘lrmsd’, ‘ilrmsd’, ‘fnat’, ‘dockq’]
short description: Which value should be used to sort the output.
long description: Which value should be used to sort the output, the output table will be sorted accordingly.
group: analysis
explevel: easy