Example of how to use regular expressions to modify FASTA headers. See: https://en.wikipedia.org/wiki/FASTA_format

modifyFastaHeaders.py

#!/usr/bin python

# We're using the sys and re modules
import sys
import re

# This is the function which will do our replacing
def modify_fasta_headers(inpath, outpath):
	# Open the in and out files
	infile = open(inpath, "r")
	outfile = open(outpath, "w")
	
	# Create the pattern we're looking for
	pattern = re.compile(r"^>[^|]+\|[^|]+\|[^|]+\|(?P<refnum>[^|]+)\|(?P<name>.+)$")
	
	# Iterate through the lines in the infile
	for line in infile.readlines():
		# Check if the line matches the pattern
		match = pattern.match(line)
		if match:
			# Extract the refnum and name from the match
			refnum = match.group("refnum")
			name = match.group("name")
			
			# Convert the refnum and name to the desired formats
			refnum = re.sub(r"^(.+)\..+$", r"\1", refnum)
			name = re.sub(r"^.*?\[(.+?)\].*$", r"\1", name).replace(" ", "_")
			
			# Write the modified line
			outfile.write(">" + refnum + "_" + name + "\n")
		else:
			# Just write the line
			outfile.write(line)
	
	# Close the files
	infile.close()
	outfile.close()

# Check if this is the main script
if __name__ == "__main__":
	# If there are no arguments (the list only has one entry: modifyFastaHeaders.py)
	if len(sys.argv) == 1:
		# Print out instructions
		print "Usage: python modifyFastaHeaders.py infile [outfile]"
		# Exit
		sys.exit(1);
	# Use the first argument as the infile
	infile = sys.argv[1]
	# If there are at least 2 arguments
	if len(sys.argv) > 2:
		# Use the second argument as the outfile
		outfile = sys.argv[2]
	else:
		# Add "_converted" to the end of the infile
		# For example:
		#   example.fasta => example_converted.fasta
		outfile = re.sub(r"(\.\w+)$", r"_converted\1", infile)
	# Run the convertions function
	modify_fasta_headers(infile, outfile)
	# Let the user know what has happened
	print infile + " converted to: " + outfile