def locus_tag2identity_best_hit_all_genomes(biodb_name,
locus,
group_name,
locus_tag2taxonomic_id_dict=""):
server, db = manipulate_biosqldb.load_db(biodb_name)
genomes = manipulate_biosqldb.get_genome_taxons_list(server, biodb_name)
#print genomes
# create empty dictionnary
genome2genome = {}
for genome1 in genomes:
if not genome1 in genome2genome.keys():
genome2genome[str(genome1)] = {}
for genome2 in genomes:
genome2genome[str(genome1)][str(genome2)] = []
try:
identity_table = np.array(
get_orthogroup_identity_table(biodb_name, group_name))
all_locus = identity_table[:, 0]
except:
genome2best_hit = {}
for genome1 in genomes:
genome2best_hit[genome1] = 0
return genome2best_hit
#print "identity table", len(identity_table)
#locus_tag2genome_name = manipulate_biosqldb.locus_tag2genome_description(server, biodb_name)
locus_tag2taxonomic_id_dict = manipulate_biosqldb.locus_tag2genome_taxon_id(
server, biodb_name)
#print "locus_tag2taxonomic_id ok"
locus_index_ref = list(all_locus).index(locus)
'''
for y in range(1, len(all_locus) + 1):
genome1 = locus_tag2taxonomic_id_dict[all_locus[y-1]]
for z in range(0, len(all_locus)):
genome2= locus_tag2taxonomic_id_dict[all_locus[z]]
genome2genome[str(genome1)][str(genome2)].append(identity_table[z, y])
'''
genome2best_hit = {}
for genome1 in genomes:
genome2best_hit[genome1] = 0
#print genome2best_hit
for target_locus in all_locus:
locus_index_target = list(all_locus).index(target_locus)
identity = identity_table[locus_index_ref, locus_index_target + 1]
target_taxon = locus_tag2taxonomic_id_dict[target_locus]
if float(identity) > genome2best_hit[str(target_taxon)]:
genome2best_hit[str(target_taxon)] = float(identity)
#print genome2best_hit
return genome2best_hit
def locus_list2presence_absence_all_genomes(locus_list, biodb_name):
server, db = manipulate_biosqldb.load_db(biodb_name)
locus_tag2seqfeature_id = manipulate_biosqldb.locus_tag2seqfeature_id_dict(
server, biodb_name)
taxon_id2description = manipulate_biosqldb.taxon_id2genome_description(
server, biodb_name)
import re
for i in taxon_id2description.keys():
taxon_id2description[i] = re.sub(" subsp\. aureus", "",
taxon_id2description[i])
taxon_id2description[i] = re.sub(", complete genome\.", "",
taxon_id2description[i])
taxon_id2description[i] = re.sub(", complete sequence\.", "",
taxon_id2description[i])
taxon_id2description[i] = re.sub("strain ", "",
taxon_id2description[i])
taxon_id2description[i] = re.sub("str\. ", "", taxon_id2description[i])
taxon_id2description[i] = re.sub(" complete genome sequence\.", "",
taxon_id2description[i])
taxon_id2description[i] = re.sub(" complete genome\.", "",
taxon_id2description[i])
taxon_id2description[i] = re.sub(" chromosome", "",
taxon_id2description[i])
taxon_id2description[i] = re.sub("Staphylococcus aureus ", "",
taxon_id2description[i])
header = 'orthogroup\t'
genomes = manipulate_biosqldb.get_genome_taxons_list(server, biodb_name)
for i in genomes:
header += taxon_id2description[i] + '\t'
final_out = header + '\n'
for i in locus_list:
#print "locus", i
seqfeature_id = locus_tag2seqfeature_id[i]
orthogroup = manipulate_biosqldb.seqfeature_id2orthogroup(
server, seqfeature_id, biodb_name)
#print "ortho", orthogroup
dico = heatmap_presence_absence(biodb_name, orthogroup)
#print "dico done..."
#print dico
out = '%s\t' % orthogroup
for i in genomes:
out += '%s\t' % dico[i]
final_out += out + '\n'
return final_out
def heatmap_presence_absence(biodb_name, group_name):
server, db = manipulate_biosqldb.load_db(biodb_name)
genomes = manipulate_biosqldb.get_genome_taxons_list(server, biodb_name)
template = ''
for i in range(0, len(genomes)):
template += '`%s`, ' % genomes[i]
template += '`%s`' % genomes[-1]
#print "template", template
sql = 'select %s from orthology_%s where orthogroup = "%s"' % (template, biodb_name, group_name)
result = [int(i) for i in server.adaptor.execute_and_fetchall(sql,)[0]]
#print result
taxon2presence_absence = {}
for x, y in zip(genomes, result):
taxon2presence_absence[x] = y
return taxon2presence_absence