ortho2pattern

ortho2pattern.py

#!/usr/bin/env python3
import csv
input = 'Orthogroups.csv'
outfile = 'phyletic_patterns.txt'

# open report file you will write to
patterns = dict()
with open(input) as csvfile:
    # columns with gene info by species are tab delimited
    reader = csv.reader(csvfile,delimiter="\t")
    header = next(reader)
    header[0] = 'Orthogroup'
    for row in reader:
        ortho_name = row[0];
        species_seen = dict()
        n_counter = 0
        for species_genes in row[1:]:
            # each of these is comma delimited
            species = header[n_counter]
            n_counter += 1
            if len(species_genes):
                for gene in species_genes.split(", "):
                    species_seen[species] = 1
                
        pattern = ",".join(sorted(species_seen.keys()))
        if pattern in patterns:
            patterns[pattern] +=1 
        else:
            patterns[pattern] = 1


with open(outfile,'w') as rpt:
# sort by abundance for a better report format
# (eg. the most abundant pattern comes first)
# https://github.com/biodataprog/Class_Examples/blob/master/Dictionaries/sort_dictionary_by_value.py
    for pattern in sorted(patterns, key=patterns.__getitem__, reverse=True):
        rpt.write("%d\t%s\n"%(patterns[pattern],pattern))