replacement_shared_snps = {cohort: [] for cohort in cohorts}
replacement_opportunities = {cohort: [] for cohort in cohorts}
for species_name in good_species_list:
sys.stderr.write("\nProcessing %s...\n" % species_name)
# First we have to enumerate QP pairs in each cohort
sys.stderr.write("Enumerating QP pairs...\n")
# all samples
all_samples = sample_order_map.keys()
# list of samples that meet coverage criteria for this species
highcoverage_samples = set(
diversity_utils.calculate_highcoverage_samples(species_name))
# list of samples that meet QP criteria for this species
haploid_samples = set(
diversity_utils.calculate_haploid_samples(species_name))
#print len(all_samples), len(highcoverage_samples), len(haploid_samples)
if len(haploid_samples) < config.within_host_min_haploid_sample_size:
continue
same_sample_idxs, same_subject_idxs, diff_subject_idxs = sample_utils.calculate_ordered_subject_pairs(
sample_order_map, all_samples)
hmp_sample_size = 0
good_species_list = parse_midas_data.parse_good_species_list()
if species != 'all':
good_species_list = [species]
else:
if debug:
good_species_list = good_species_list[:3]
# header for the output file.
record_strs = []
for species_name in good_species_list:
sys.stderr.write("Loading samples...\n")
# Only plot samples above a certain depth threshold that are confidently phaseable.
snp_samples = diversity_utils.calculate_highcoverage_samples(
species_name, min_coverage=min_coverage)
if len(snp_samples) < 2:
continue
sys.stderr.write("found %d samples\n" % len(snp_samples))
# Analyze SNPs, looping over chunk sizes.
# Clunky, but necessary to limit memory usage on cluster
# Load SNP information for species_name
sys.stderr.write("Loading SNPs for %s...\n" % species_name)
snps = []
snp_map = {} # contig: list of locations map