def dstruc_loadin(self, curs):
"""
03-09-05
get the context from mcl_table via linking through mcl_id of p_gene_table
context_dict is set
"""
from codense.common import get_known_genes_dict, get_go_no2go_id, get_go_no2name, get_gene_no2gene_id
self.known_genes_dict = get_known_genes_dict(curs)
self.go_no2go_id = get_go_no2go_id(curs)
self.go_no2go_name = get_go_no2name(curs)
self.gene_no2gene_id = get_gene_no2gene_id(curs)
sys.stderr.write("Setting up gene_prediction_dict...")
# setup self.gene_prediction_dict
curs.execute(
"select p.gene_no, p.go_no, p.is_correct, p.is_correct_l1, p.is_correct_lca, m.vertex_set\
from %s p, %s g, %s m where g.p_gene_id=p.p_gene_id and m.mcl_id=p.mcl_id"
% (self.gene_table, self.table, self.mcl_table)
)
rows = curs.fetchall()
for row in rows:
gene_no = row[0]
if self.type == 2 and gene_no not in self.known_genes_dict:
# I only want the known genes, but this gene is unknown
continue
elif self.type == 3 and gene_no in self.known_genes_dict:
# i only want the unknown genes, but this gene is known
continue
go_no = row[1]
is_correct = row[2]
is_correct_l1 = row[3]
is_correct_lca = row[4]
vertex_set = row[5][1:-1].split(",")
vertex_set = map(int, vertex_set)
item = function_struc()
item.is_correct = is_correct
item.is_correct_l1 = is_correct_l1
item.is_correct_lca = is_correct_lca
# context_dict is a set
item.context_dict = Set(vertex_set)
if gene_no not in self.gene_prediction_dict:
self.gene_prediction_dict[gene_no] = gene_prediction()
self.gene_prediction_dict[gene_no].p_functions_struc_dict[go_no] = item
else:
self.gene_prediction_dict[gene_no].p_functions_struc_dict[go_no] = item
sys.stderr.write("Done\n")
"""
def dstruc_loadin(self, curs):
"""
03-14-05
remove the distance loading part
"""
sys.stderr.write("Loading Data STructure...\n")
from codense.common import get_known_genes_dict, get_go_no2go_id,\
get_go_no2term_id, get_go_no2depth, get_go_term_id2go_no, \
get_go_term_id2depth
self.known_genes_dict = get_known_genes_dict(curs)
self.go_no2go_id = get_go_no2go_id(curs)
self.go_no2term_id = get_go_no2term_id(curs)
self.go_no2depth = get_go_no2depth(curs)
self.go_term_id2go_no = get_go_term_id2go_no(curs)
self.go_term_id2depth = get_go_term_id2depth(curs)
sys.stderr.write("Done\n")
def parse_cluster_fname(self, curs, cluster_fname, gim_inputfname, cluster_id_set, schema_instance):
"""
01-24-06
a lot of analogy to codense2db.py's run()
"""
sys.stderr.write("Parsing cluster_fname: %s ...\n"%os.path.basename(cluster_fname))
codense2db_instance = codense2db()
codense2db_instance.create_tables(curs, schema_instance.splat_table, \
schema_instance.mcl_table, schema_instance.pattern_table)
gene_id2gene_no = get_gene_id2gene_no(curs)
gene_no2incidence_array = get_gene_no2incidence_array(gim_inputfname, gene_id2gene_no)
known_gene_no2go_no_set = get_known_genes_dict(curs)
counter = 0
real_counter = 0
cluster_id2properties = {} #additional properties for prediction_pair2instance
reader = csv.reader(open(cluster_fname, 'r'), delimiter='\t')
for row in reader:
counter += 1
#only those who are in cluster_id_set
if counter in cluster_id_set: #cluster_id starts from 1
cluster_list = codense2db_instance.fimbfs_parser(row, gene_no2incidence_array, curs)
for cluster in cluster_list:
real_counter += 1
cluster.unknown_gene_ratio = codense2db_instance.calculate_unknown_gene_ratio(cluster.vertex_set, \
known_gene_no2go_no_set)
cluster.cluster_id = counter #line number is the cluster_id
codense2db_instance.db_submit(curs, cluster, schema_instance.pattern_table)
cluster_id2properties[cluster.cluster_id] = [cluster.connectivity, cluster.unknown_gene_ratio, cluster.vertex_set]
if real_counter==len(cluster_id_set):
#all relevant clusters have been got, ignore remaining clusters
break
if self.report and counter%2000==0:
sys.stderr.write("%s%s/%s"%('\x08'*20, counter, real_counter))
if self.report:
sys.stderr.write("%s%s/%s"%('\x08'*20, counter, real_counter))
del reader
sys.stderr.write("Done.\n")
return cluster_id2properties
