def load_centroid_gene_map(desired_species_name=None):
if desired_species_name==None:
from parsers import parse_midas_data
desired_speciess = parse_midas_data.parse_good_species_list()
else:
desired_speciess = [desired_species_name]
for desired_species_name in desired_speciess:
# First load reference genes
reference_genes = load_reference_genes(desired_species_name)
gene_info_file = gzip.open("%span_genomes/%s/gene_info.txt.gz" % (config.midas_directory, desired_species_name), 'r')
gene_info_file.readline() # header
centroid_gene_map = {}
for line in gene_info_file:
items = line.split("\t")
gene_id = items[0].strip()
centroid_id = items[3].strip()
if centroid_id not in centroid_gene_map:
centroid_gene_map[centroid_id] = centroid_id
if (gene_id in reference_genes) and (centroid_id not in reference_genes):
centroid_gene_map[centroid_id] = gene_id
gene_info_file.close()
return centroid_gene_map
parser.add_argument("--species", help="Name of specific species to run code on", default="all")
args = parser.parse_args()
debug = args.debug
chunk_size = args.chunk_size
species=args.species
file = gzip.GzipFile(intermediate_filename,"w")
# Load subject and sample metadata
sys.stderr.write("Loading sample metadata...\n")
subject_sample_map = parse_HMP_data.parse_subject_sample_map()
sys.stderr.write("Done!\n")
# get a list of specis to run this script on.
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)
from parsers import parse_midas_data
if len(sys.argv) < 3:
sys.stderr.write(
"Usage: python loop_over_species_wrapper.py all|debug|species command...\n"
)
sys.exit(1)
# First argument is either 'all', 'debug', or a species name
debug_flag = ""
if sys.argv[1] == 'debug':
species_names = [parse_midas_data.debug_species_name]
debug_flag = "--debug"
elif sys.argv[1] == 'all':
species_names = parse_midas_data.parse_good_species_list()
else:
good_species_names = parse_midas_data.parse_good_species_list()
species_names = []
pattern = sys.argv[1]
for species_name in good_species_names:
if species_name.startswith(pattern):
species_names.append(species_name)
# Remaining arguments are command to run, with species name appended as last argument
command = " ".join(sys.argv[2:])
sys.stderr.write("Running command: %s\n" % command)
sys.stderr.write("for %d species...\n\n" % len(species_names))
for species_name in species_names:
f = float(items[1])
gene_freq_map[gene_name] = f
file.close()
return gene_freq_map
# Actually calculate the core genes
if __name__ == '__main__':
from parsers import parse_midas_data
os.system('mkdir -p %s' % core_genes_directory)
os.system('mkdir -p %s' % external_core_genes_directory)
pangenome_species = parse_midas_data.parse_good_species_list()
cmin = config.core_genome_min_copynum
cmax = config.core_genome_max_copynum
shared_cmin = config.shared_genome_min_copynum
min_good_fraction = config.core_genome_min_prevalence
min_coverage = 5 # (for assessing core genome, we'll use a lower coverage value than when we look at real changes)
output_filename = default_core_gene_filename
output_file = gzip.GzipFile(output_filename, "w")
stringent_output_filename = default_stringent_core_gene_filename
stringent_output_file = gzip.GzipFile(stringent_output_filename, "w")
shared_output_file = gzip.GzipFile(default_shared_gene_filename, "w")
