[UniProt Challenge] PSI-BLAST & PSSM : Protein representation

 

import subprocess
import os


# Local Psi Blast installation path
path_to_psiblast = 'C:\\Program Files\\NCBI\\blast-2.7.1+\\bin\\psiblast.exe'


# Path to Proteins in Fasta format
fasta_path = 'processed_fastas/mouse_train/'


from os import listdir
from os.path import isfile, join
onlyfiles = [f for f in listdir(fasta_path) if isfile(join(fasta_path, f))]


# psiblast -query A0JNU3.fasta -db swissprot/swissprot -num_iterations 3 -evalue 0.001 -num_threads 8 -save_each_pssm -out_ascii_pssm A0JNU3.pssm
for i in onlyfiles:
    query_fasta = fasta_path + i

 # Output filename for each PSSM
    output_pssm = 'pssm/mouse_train/' + i + '.pssm'

 # Call the sub process with proper arguments
    subprocess.call([path_to_psiblast, '-query', query_fasta, '-db', 'uniref50/uniref50.fasta', '-num_iterations', '3', '-evalue', '0.001', '-num_threads', '8', '-out_ascii_pssm', output_pssm])

Python Sub Process Local Psi Blast PSSM Generation from FASTA in Directory using Uniref50 Database in Pycharm (quickgrid.blogspot.com)

Python Sub Process Local Psi Blast PSSM Generation from FASTA in Directory using Uniref50 Database in Pycharm

/ncbi-blast-2.9.0+/bin$ psiblast \
-db /data/BlastDB/uniref50.fasta \
-query Test_1999.fasta \
-evalue 0.001 \
-out /add_valid_path/Test.txt \
-out_ascii_pssm /add_valid_path/seq.1.pssm \
-num_threads 8 \
–num_iterations 2

- PSSM 은 여전히 강력한 피쳐임

- 길이, 다른 시퀀스 특징을 고려해서 파라미터를 조정해야 함.