PSSMGen

Generates consistent PSSM and/or PDB files for protein-protein complexes

Install

1. Make sure BLAST is installed and its database is available on your machine. Otherwise, install BLAST and download its databases by following the BLAST guide.
2. Install the PSSMgen by pip install PSSMGen.

Requirements for file structures and names

PSSMGen is geared toward computing the pssm files for all models of a particular protein-protein complex.

File structures

This tool assumes your files have following structure:

 workdir
 |_ pdb
 |_ fasta
 |_ pssm_raw
 |_ pssm
 |_ pdb_nonmatch

• workdir is your working directory for one specific protein-protein complex.
• pdb folder must contain the PDB files.
• fasta folder contains the protein sequence FASTA files. The code can generate the FASTA files by extracting sequences from the pdb file , or you can manually create this folder and put customed FASTA files there.
• pssm_raw folder stores the PSSM files. The code can automatically generate them, or you can manually create this folder and put customed PSSM files there.
• pssm folder stores consistent PSSM files, whose sequences are aligned with those of PDB files. This folder and its files are created automatically.
• pdb_nonmatch folder stores the inconsistent PDB files, while the related consistent PDB files are in the pdb folder. This folder and its files are created automatically.

File names

The code assumes you follow the naming rules for different file types:

• PDB files: caseID_*.chainID.pdb
• FASTA files: caseID.chainID.fasta
• PSSM files: caseID.chainID.pssm, caseID_*.chainID.pdb.pssm

Examples

Get consistent PSSM and PDB files for a specific protein-protein complex

Here is an example for complex 7CEI:

The file structure and input files should look like:

7CEI
└── pdb
    ├── 7CEI_1w.pdb
    ├── 7CEI_2w.pdb
    └── 7CEI_3w.pdb

from pssmgen import PSSM

# initiate the PSSM object
gen = PSSM(work_dir='7CEI')

# set psiblast executable, database and other psiblast parameters (here shows the defaults)
gen.configure(blast_exe='/home/software/blast/bin/psiblast',
            database='/data/DBs/blast_dbs/nr_v20180204/nr',
            num_threads = 4, evalue=0.0001, comp_based_stats='T',
            max_target_seqs=2000, num_iterations=3, outfmt=7,
            save_each_pssm=True, save_pssm_after_last_round=True)

# generates FASTA files
gen.get_fasta(pdb_dir='pdb', chain=['A','B'], out_dir='fasta')

# generates PSSM
gen.get_pssm(fasta_dir='fasta', out_dir='pssm_raw', run=True)

# map PSSM and PDB to get consisitent files
gen.map_pssm(pssm_dir='pssm_raw', pdb_dir='pdb', out_dir='pssm', chain=['A','B'])

# write consistent files and move
gen.get_mapped_pdb(pdbpssm_dir='pssm', pdb_dir='pdb', pdbnonmatch_dir='pdb_nonmatch')

The code will automatically create fasta, pssm_raw, pssm and pdb_nonmatch folders and related files.

Using customed FASTA files

You can use customed FASTA files intead of extracting them from PDB file.

The file structure and input files should look like

7CEI
└── fasta
    ├── 7CEI.A.fasta
    └── 7CEI.B.fasta

from pssmgen import PSSM

# initiate the PSSM object
gen = PSSM('7CEI')

# set psiblast executable, database
gen.configure(blast_exe='/home/software/blast/bin/psiblast',
            database='/data/DBs/blast_dbs/nr_v20180204/nr')

# generates PSSM
gen.get_pssm()

Using customed PSSM files

You can also use avaliable PSSM files intead of calculating them again.

The file structure and input files should look like

7CEI
├── pdb
│   ├── 7CEI_1w.pdb
│   ├── 7CEI_2w.pdb
│   └── 7CEI_3w.pdb
└── pssm_raw
    ├── 7CEI.A.pssm
    └── 7CEI.B.pssm

from pssmgen import PSSM

# initiate the PSSM object
gen = PSSM('7CEI')

# map PSSM and PDB to get consisitent files
gen.map_pssm()

# write consistent files and move
gen.get_mapped_pdb()