def run(self):
"""
09-08-05
--db_connect()
--org_short2long()
--org2tax_id()
--setup_acc2gene_id()
if self.new_table
--create_output_table()
--parse_input_filename()
"""
(conn, curs) = db_connect(self.hostname, self.dbname)
long_organism = org_short2long(self.organism)
tax_id_set = Set([org2tax_id(long_organism)])
MdbId2GeneId_instance = MdbId2GeneId()
acc2gene_id = MdbId2GeneId_instance.setup_acc2gene_id(
self.acc_file, tax_id_set)
if self.new_table:
self.create_output_table(curs, self.output_table)
self.parse_input_filename(curs, self.input_filename, self.output_table, acc2gene_id,\
org2tax_id(long_organism), self.up_length, self.comment, long_organism, self.type)
if self.commit:
curs.execute("end")
def run(self):
"""
09-19-05
rewrite
--db_connect()
--get_global_gene_id2gene_no()
--org2tax_id()
--get_gene_id2mt_no_list()
--return_gene_id_set()
--submit()
"""
(conn, curs) = db_connect(self.hostname, self.dbname, self.schema)
gene_id2gene_no = get_global_gene_id2gene_no(curs, self.organism)
tax_id = org2tax_id(self.organism)
"""
#01-14-06 comment it out for future
gene_no2tf_set = get_gene_no2tf_set(curs) #12-15-05 just yeast.
#12-15-05 convert gene_no(integer) into gene_id(string)
gene_id2mt_no_list = {}
for gene_no, tf_set in gene_no2tf_set.iteritems():
gene_id2mt_no_list[repr(gene_no)] = list(tf_set)
"""
gene_id2mt_no_list = get_gene_id2mt_no_list(tax_id)
gene_id_set = self.return_gene_id_set(self.dir, gene_id2gene_no, self.min_frequency)
self.submit(curs, output_table, gene_id_set, gene_id2gene_no, gene_id2mt_no_list)
if self.needcommit:
conn.commit()
def run(self):
"""
07-31-05
"""
if not os.path.isdir(self.outputdir):
os.makedirs(self.outputdir)
tax_id = org2tax_id(self.organism)
mapping_dict = get_unigene2gene_list(self.mapping_file, tax_id)
sys.stderr.write("\tTotally, %d files to be processed.\n"%len(self.files))
for f in self.files:
sys.stderr.write("%d/%d:\t%s"%(self.files.index(f)+1,len(self.files),f))
self.transform_one_file(f, self.delimiter, self.outputdir, mapping_dict, self.type)
sys.stderr.write("\n")
def run(self): """ 09-28-05 12-19-05 use class_list and output_fname_list to ease program writing 12-30-05 fix a bug in indexing darwin_instance_list 2006-09-25 2007-02-08 add context_prediction_csv_format """ tf_darwin_ofname = os.path.join(self.output_dir, '%s.tf.darwin'%self.cluster_bs_table) cluster_darwin_ofname = os.path.join(self.output_dir, '%s.cluster.darwin'%os.path.basename(self.input_fname)) prediction_darwin_ofname = os.path.join(self.output_dir, '%s.prediction.darwin'%os.path.basename(self.input_fname)) pattern_darwin_ofname = os.path.join(self.output_dir, '%s.pattern.darwin'%self.pattern_table) if not os.path.isdir(self.output_dir): os.makedirs(self.output_dir) conn, curs = db_connect(self.hostname, self.dbname, self.schema) tax_id = org2tax_id(self.organism) #gene_no2id = get_gene_no2gene_id(curs) #Watch, if unigene, should use this. gene_id2symbol = get_gene_id2gene_symbol(curs, tax_id) gene_id2symbol = self.replace_prime_in_gene_id2symbol(gene_id2symbol) #01-26-06 #gene_no2symbol = dict_transfer(gene_no2id, gene_id2symbol) #Jasmine wants the gene symbol 09-28-05 #gene_id is integer in gene.gene table and same as gene_no, so just use it. go_no2name = get_go_no2name(curs) #09-28-05 Jasmine wants the go_name, not go_id #2006-09-25 use gene_id2symbol to replace mt_no2tf_name #mt_no2tf_name = get_mt_no2tf_name() mt_no2tf_name = gene_id2symbol class_list = [tf_darwin_format, cluster_darwin_format, prediction_darwin_format, pattern_darwin_format, context_prediction_csv_format] context_prediction_csv_fname = os.path.join(self.output_dir, '%s.context.csv'%self.input_fname) output_fname_list = [tf_darwin_ofname, cluster_darwin_ofname, prediction_darwin_ofname, pattern_darwin_ofname, context_prediction_csv_fname] darwin_instance_list = [] for i in range(len(self.running_bit)): if self.running_bit[i] == '1': darwin_instance_list.append(class_list[i](self.hostname, self.dbname, self.schema, self.pattern_table,\ self.cluster_bs_table, self.input_fname, self.lm_bit, self.acc_cut_off, \ output_fname_list[i], gene_id2symbol, go_no2name, mt_no2tf_name, debug, report)) #2006-09-25 current_pos = len(darwin_instance_list)-1 #12-30-05 darwin_instance_list[current_pos].start() for i in range(len(darwin_instance_list)): darwin_instance_list[i].join()
def run(self):
"""
11-15-05
correct a bug related to self.size
2006-08-27
if sequence is empty, ignore it.
2006-11-27
add running_type
"""
if not os.path.isdir(self.folder):
os.makedirs(self.folder)
(conn, curs) = db_connect(self.hostname, self.dbname, self.schema)
if self.running_type == 1:
curs.execute(
"DECLARE crs CURSOR FOR SELECT id, sequence from prom_seq \
where sequence is not null and strpos(chromosome, 'random')=0 and organism='%s'"
% self.organism)
#09-14-05 not null sequence and no 'random' in chromosome
elif self.running_type == 2:
curs.execute(
"DECLARE crs CURSOR FOR SELECT r.id, r.sequence from sequence.raw_sequence r, sequence.annot_assembly a\
where r.acc_ver=a.acc_ver and a.tax_id=%s" % org2tax_id(self.organism))
#2006-11-27 from a specific tax_id, the later condition will guarantee that sequence is not empty
else:
sys.stderr.write("Unsupported running_type: %s\n" %
self.running_type)
sys.exit(3)
curs.execute("fetch %s from crs" % self.size)
rows = curs.fetchall()
counter = 0
sys.stderr.write("Starting to output...\n")
while rows:
output_file = os.path.join(self.folder,
'%s%s' % (self.prefix, counter))
of = open(output_file, 'w')
for row in rows:
id, sequence = row
if sequence:
of.write('>%s\n%s\n' % (id, sequence))
del of
counter += 1
if self.report:
sys.stderr.write('%s%s' % ('\x08' * 20, counter))
curs.execute("fetch %s from crs" % self.size)
rows = curs.fetchall()
del conn, curs
sys.stderr.write("Done.\n")
def run(self):
"""
09-18-05
09-30-05
way of calling get_mt_id2no() changed
get gene_id_dict is a thread now, to speed up
"""
(conn, curs) = db_connect(self.hostname, self.dbname)
tax_id = org2tax_id(self.organism)
tax_id_set = Set([tax_id])
get_gene_id_dict_instance = get_gene_id_dict(self.acc_file, tax_id_set)
get_gene_id_dict_instance.start()
mt_id2no = get_mt_id2no(curs, self.matrix_table)
mt_no2matches = self.get_mt_no2matches(curs, self.input_table, self.prom_seq_table, self.top_number, self.organism, mt_id2no)
get_gene_id_dict_instance.join() #must wait it to finish before going on, need gene_id_dict
self.dump2output_table(curs, self.output_table, mt_no2matches, get_gene_id_dict_instance.gene_id_dict, tax_id)
if self.commit:
curs.execute("end")
def run(self):
"""
11-15-05
correct a bug related to self.size
2006-08-27
if sequence is empty, ignore it.
2006-11-27
add running_type
"""
if not os.path.isdir(self.folder):
os.makedirs(self.folder)
(conn, curs) = db_connect(self.hostname, self.dbname, self.schema)
if self.running_type==1:
curs.execute("DECLARE crs CURSOR FOR SELECT id, sequence from prom_seq \
where sequence is not null and strpos(chromosome, 'random')=0 and organism='%s'"%self.organism)
#09-14-05 not null sequence and no 'random' in chromosome
elif self.running_type==2:
curs.execute("DECLARE crs CURSOR FOR SELECT r.id, r.sequence from sequence.raw_sequence r, sequence.annot_assembly a\
where r.acc_ver=a.acc_ver and a.tax_id=%s"%org2tax_id(self.organism))
#2006-11-27 from a specific tax_id, the later condition will guarantee that sequence is not empty
else:
sys.stderr.write("Unsupported running_type: %s\n"%self.running_type)
sys.exit(3)
curs.execute("fetch %s from crs"%self.size)
rows = curs.fetchall()
counter = 0
sys.stderr.write("Starting to output...\n")
while rows:
output_file = os.path.join(self.folder, '%s%s'%(self.prefix, counter))
of = open(output_file, 'w')
for row in rows:
id, sequence = row
if sequence:
of.write('>%s\n%s\n'%(id,sequence))
del of
counter += 1
if self.report:
sys.stderr.write('%s%s'%('\x08'*20, counter))
curs.execute("fetch %s from crs"%self.size)
rows = curs.fetchall()
del conn, curs
sys.stderr.write("Done.\n")