iScore
A framework and predictor based on support vector machine and random walk graph kernel for scoring protein-protein interfaces.
Generates consistent PSSM and PDB files for protein-protein complexes
PSSMGen is geared toward computing the pssm files for all models of a particular protein-protein complex. This tool assumes your files have following structure:
caseID
|_ pdb
|_ fasta
|_ pssm_raw
|_ pssm
|_ pdb_raw
caseID is the ID of protein-protein complex, e.g. PDB ID 1AK4 or CAPRI target ID T161. Only the pdb dir must exist at run time, and it must contain the PDB files of the models generated by HADDOCK. Based on these PDB files, the code will extract protein sequences and generate the FASTA files (stored in the fasta subdir). It will then use the BLAST tool psiblast to compute PSSM (stored in pssm_raw subdir). Finally the sequences from PSSM files and PDB files will be aligned and then the consistent PSSM files and/or PDB files will generated. The consistent PSSM files are stored in pssm subdir. If applicable, the consistent PDB files are stored in pdb subdir, while the original non-consistent PDB files are moved to pdb_raw subdir. You can then use PDB files from pdb and PSSM files from pssm for further analysis.
Scoring 3D protein-protein interaction models using deep learning
A framework and predictor based on support vector machine and random walk graph kernel for scoring protein-protein interfaces.
Fast and versatile Python package that leverages SQL queries to parse, manipulate and process biomolecular structure files. The structure files should be in the PDB format and are available on www.rcsb.org.