getRegLoc

#!/usr/bin/python
'''
Obtain the orthologs of a series of regulators, producing a series of "regloc" files

regloc file:
locus_tag   REG_NAME

pangenome tab file:
orth_id    locus_tag

target locus file:
prot_id locus_tag   [dna_id replicon]
'''

def getOptions():
    import argparse

    # create the top-level parser
    description = ('Obtain the orthologs of a series of regulators, producing a series of "regloc" files')
    parser = argparse.ArgumentParser(description = description)
    parser.add_argument('inregloc', action='store', 
                            help='Seed regloc file')
    parser.add_argument('pangenome', action='store',
                            help='Pangenome tab file')
    parser.add_argument('locusfile', action='store', 
                            help='Target locus file')
    return parser.parse_args()
    
options = getOptions()

r = {}
for l in open(options.inregloc):
    locus, reg = l.strip().split('\t')
    r[reg] = r.get(reg, [])
    r[reg].append(locus)
    
p = {}
o = {}
for l in open(options.pangenome):
    if l.strip().startswith('#'):continue
    orth, protein = l.strip().split('\t')
    p[orth] = p.get(orth, set())
    p[orth].add(protein)
    o[protein] = orth
    
g = set()
for l in open(options.locusfile):
    s = l.strip().split('\t')
    g.add(s[1])

# Cycle through each regulator
for reg, locuses in r.iteritems():
    matches = 0
    for locus in locuses:
        orth = o[locus]
        
        # Is my target organism in this group?
        # It may have some paralogs!
        for protein in p[orth]:
            if protein in g:
                matches += 1
                print '\t'.join([protein, reg])
                
    if matches == 0:
        print '\t'.join(['NA', reg])