FASTAfilter.py

#!/usr/bin/env python
import re, sys
import argparse

# This script was written by David Molnar
# and downloaded from
# http://dm516.user.srcf.net/?p=314
# 2016-04-13

parser = argparse.ArgumentParser(description='From a FASTA-file with multiple >entries, filter by sequence ids using a regex.')
parser.add_argument('regex',
                   help="Regex to filter entry ids, e.g. 'chr[1-4]'. Note that the id does not contain the initial > character.")
parser.add_argument('infile',
                   help='A FASTA input file, usually with multiple entries.') 
parser.add_argument('outfile', 
                   help='The new file with only the matching entries.')                                      

args = parser.parse_args()

reader = open(args.infile, "rb")
outfile = open(args.outfile, "w")
filter = re.compile(args.regex)
title=""
buffer=""

def useBuffer(title, buffer):
	if filter.match(title):
		print "Used: " + title
		outfile.write(">"+title+"\n")
		outfile.write(buffer)
		outfile.write("\n")
	else:
		print "Skipped: "+title
		 

while 1:
	line = reader.readline()
	if not line:
		#This is the end
		if title:
			#use the last title and buffer...
			useBuffer(title, buffer)
		break
	
	line = line.strip()

	if line == "":
		continue

	if line.startswith(">"):
		#this is a title -- starts a new fasta block		 
		if buffer:
			#use the last title and buffer somehow...
			useBuffer(title, buffer)
		
		buffer = ""	
		title = line [1:]
		continue
		
	if title:		
		buffer += line
			
reader.close()			  
outfile.close()