Hi all,
I am trying to run a tblastn search of nt locally using blast+2.2.25, and was wondering if anyone new if it's going as fast as it should - I have 1000's of queries to get through!
I should mention - at the moment I am blasting via a python script which blasts in batches batches that i define. I've tried 5 sets of 20-query batches, which took and average of 20 mins to run..
whats the slowest part of the blast search? what could i do to speed it up.. the ideal runtime would be 15-30 seconds per query.
Im using 8 cores by the way, with 12 GB of RAM. and ive changed -num_threads to 16, also excluded some gi's..any other ideas?Thanks to all of you that answered! it's my first post and I'm quite touched that i got this much support from strangers! inspires me to contribute as much as i can!
#!/usr/bin/python
#BLASTing in a way that doesn't crash the webserver!
from time import sleep
from Bio import SeqIO
from Bio.Blast import NCBIWWW
import os
from Bio.Blast.Applications import NcbitblastnCommandline
import time
mult = input('How many queries shall I BLAST each time time?')
recfile= input('\n\nI need the file name containging your genpept records\n\nPlease enclose entry with single quotes:')
########################
def numbsuffix(d):
return 'th' if 11<=d<=13 else {1:'st',2:'nd',3:'rd'}.get(d%10, 'th')
####################### ...