PISA (Protein Interfaces, Surfaces and Assemblies)

Description

Note: This version of PISA is identical to that used in the interactive tool PISA released in the latest version of the CCP4 Software Suite, version 9.0.

PISA was developed for the investigation of macromolecular interfaces (e.g. between proteins and nucleic acids) and to predict the most probable quaternary structures or assemblies.

In the latest version of PISA (CCP4 v 9.0), we have added an option to perform only assembly interface analysis and to opt out of the calculation and prediction of assemblies. The data generated by PISA has been expanded to include more information on individual interfaces, including more types of atom-atom interactions (see xml section). PDBe uses this option in PISA to efficiently calculate macromolecular interaction data, which is integrated with functional annotations from PDBe-KB and displayed on PDBe and PDBe-KB pages.

Compiling and running PISA

1. Download the source code

git clone https://github.com/PDBe-KB/pisa
cd pisa

2. Either export the environment variable $SRCDIR to point the compiler to the correct directory. E.g.:

export SRCDIR=/path/to/your/pisa

Or edit the compile.sh file in the root directory of the repository (.../pisa/compile.sh) directly

Add the complete path to the pisa directory:

srcdir=/path/to/your/pisa

4. Compile PISA

chmod +x compile.sh
./compile.sh

Configuration

Running PISA requires a configuration file. An example file is available here.

The paths for the following items have to be configured:

• DATA_ROOT
• SRS_DIR
• MOLREF_DIR
• PISTORE_DIR

Usage

After compiling, the executable for PISA will be .../pisa/build/pisa.

To run PISA analysis for an input file (PDB or mmCIF file), the following command:

pisa [name] -analyse [coorfile] [options] [cfg]

[name]: (required) Session name. The results will be stored in a directory named after the session name.

[coorfile] : (required) Path to coordinates file (PDB or mmCIF)

[options] : Optional processing keys (see below)

[cfg]: (required) Path to configuration file (it must be provided unless it is already set as environmental variable in PISA_CONF_FILE). Note that this always has to be the last parameter. A default file (pisa_cfg) is provided in the repository root.

Options

Type ./build/pisa to view all the options and their explanations.

AS IS flag

PISA has the option to only perform the analysis of interfaces for a given input assembly, omitting the calculation of probable assemblies

pisa [name] -analyse [coorfile] --as-is [cfg]

If the flag --as-is is used, PISA will analyse the interfaces in the given input assembly and omit the prediction of assemblies from crystallographic data.

XML file

PISA generates the results in binary format. The data has to be converted into a human-readable XML format using the following command:

pisa name -xml {interfaces|assemblies} [cfg] > outputfile.xml

We can optionally add the flag --as-is to generate an xml file containing information about the interfaces of the given assembly, omitting all other parameters/variables related to the prediction of assemblies.

pisa name -xml {interfaces|assemblies} --as-is > outputfile.xml

Note that the name parameter has to match an existing session name. For example, if PISA was used to analyse a PDB structure and the name of that session was analyise_3bow, then to convert the binary to XML, the name parameter has to be analyise_3bow as well.

Also note that interfaces and assemblies are the exact terms expected by the command.

Interfaces:

The xml output file for interfaces has the following information:

For each interface:

• id : Interface ID
• int_area : Area of interface (A^2)
• Int_solv_en : Solvation energy (kcal/mol)
• pvalue : Probability that solvation energy gain for interface atom may be greater than binding energy
• stab_en : Stabilisation energy (Kcal/mol)

For each bond type:

• h-bonds : Hydrogen bonds
• salt-bridges : Salt bridges
• ss-bonda : Disulfide bonds
• cov-bonds : Covalent Bonds
• other-bonds : Other interface contacts within distance cutoff 4.0 A

Details (for each bond of each type):

• n_bonds : number of bonds
• res-1 : name of residue 1
• chain-1 : name chain 1 (auth)
• label_asym_id-1 : name chain (asym) 1
• orig_label_asym_id-1 : Asym chain identifier in the original model file
• pdbx_sifts_xref_db_acc-1 : Uniprot accession number
• pdbx_sifts_xref_db_num-1 : Sequence position of the UniProt entry that corresponds to the residue mapping
• seqnum-1 : Sequence number atom 1 (auth)
• label_seqnum-1 : Sequence number atom 1 (label)
• atname-1 : atom 1 name
• Inscode-1 : insertion code of 1st linked atom
• res-2 : name of Residue 2
• chain-2 : name chain 2 (auth)
• label_asym_id-2 : name chain (asym)
• orig_label_asym_id-2 : Asym chain identifier in the original model file
• pdbx_sifts_xref_db_acc-2 : Uniprot accession number
• pdbx_sifts_xref_db_num-2 : Sequence position of the UniProt entry that corresponds to the residue mapping
• seqnum-2 : Sequence number atom 1
• label_seqnum-2 : Sequence number atom 2 (label)
• atname-2 : atom 1 name
• Inscode-2 : Insertion code of 2nd linked atom
• Dist : Bond distance (A)

Residue list (per interface ID)

For each Residue :

• ser_no : residue numbering
• name : residue name
• seq_num : sequence number
• label_seq_num : sequence number (label)
• ins_code : insertion code residue
• bonds :
• solv_en : Solvation energy effect (kcal/mol)
• asa : accessible surface area (A^2)
• bsa : Buried surface area (A^2)

If the calculation of the probable assemblies is performed, this information is shown:

• ccs score : Complex Formation Significance Score. CSS ranges from 0 to 1 as interface relevance to complex formation increases.
• Symmetry operations
• Unit cell parameters [cell unit rxx,rxy,rxz,ryx,ryy,ryz,ty,rzx,rzy,rzz,tz]

Assemblies:

The xml output file for assemblies has the following information:

• diss_energy : maximal free energy of dissociation (kcal/mol)
• diss_energy0 : ground-level free energy of dissociation (kcal/mol)
• asa : accessible surface area (A^2)
• bsa : buried surface area (A^2)
• entropy : entropy change at dissociation
• Entropy_0 : ground-level entropy change at dissociation
• diss_area : dissociation Interface Area (A^2)
• int_energy : solvation Energy Gain (kcal/mol)
• n_diss : number of dissociating parts
• Formula : assembly formula
• composition : assembly composition

If the calculation of the probable assemblies is performed, this additional information is shown:

• Assessment : contains yes for stable assemblies and no for those that are likely to dissociate in solution. This classification is based on the value of free energy of dissociation.
• Score : Assembly score based on assembly size (number of asymmetric units). It indicates if the assembly appears to be stable in solution
• Symmetry operations
• Symmetry ID
• Unit cell parameters [cell unit rxx,rxy,rxz,ryx,ryy,ryz,ty,rzx,rzy,rzz,tz]

Run in Docker

A Docker image is available for PISA. To run PISA in Docker, use the following command:

# Run the analysis
docker run -v $PWD:/data pdbegroup/pisa pisa test-session -analyse /data/example_data/6gve.cif /usr/share/pisa/setup/pisa_cfg_tmp

# Get the interfaces
docker run -v $PWD:/data pdbegroup/pisa pisa test-session -xml interface /usr/share/pisa/setup/pisa_cfg_tmp

The above command will create a session named test-session, run PISA on the provided example data (6gve.cif) and store the results in the session directory. We mount the current directory to the /data directory in the Docker container. So the sessions would be stored in the sessions directory in the current directory.

Acknowledgements

This code is based on CCP4 PISA and has been developed through a collaboration between EMBL-EBI and CCP4.