More patterns for matching unclassified sequences

unclassified.py

#!/usr/bin/env python

import re
import sys

rexp = re.compile(r'|'.join([
            r'\bactinomycete\b',
            r'\bcrenarchaeote\b',
            r'\bculture\b',
            r'\bchimeric\b',
            r'\bcyanobiont\b',
            'degrading',
            r'\beuryarchaeote\b',
            'disease',
            r'\b[cC]lone',
            r'\bmethanogen(ic)?\b',
            'planktonic',
            r'\bplanctomycete\b',
            r'\bsymbiote\b',
            r'\btransconjugant\b',
            r'^[a-z]', # need to look for false positives
            r'^[a-zA-Z]+\s+[a-zA-Z]*\d' # digit in second word
            ]))

for line in sys.stdin:
    if rexp.search(line.split(None, 3)[-1]):
        sys.stdout.write(line)