gmap_SR_cmd = (
gmap_path
+ " "
+ gmap_option
+ " "
+ " -D "
+ gmap_folder
+ " -d "
+ gmap_index
+ " "
+ SR_pathfilename
+ " > "
+ output_path
+ SR_filename
+ ".psl"
)
print gmap_SR_cmd
log_command(gmap_SR_cmd)
# notice we are not skipping any lines in the PSL file output by gmap (hence the zero in the following command)
bestblat_SR_cmd = (
python_path + " " + bin_path2 + "blat_best.py " + output_path + SR_filename + ".psl 0 > " + output_pathfilename
)
log_command(bestblat_SR_cmd)
rm_SRpsl_cmd = "rm " + output_path + SR_filename + ".psl "
print rm_SRpsl_cmd
log_command(rm_SRpsl_cmd)
def print_run(cmd, ignorefail=False):
print cmd
print ""
log_command(cmd, ignorefail)
ext_ls=[]
j=0
k=0
i=0
while i <Nthread1:
ext_ls.append( '.' + string.lowercase[j] + string.lowercase[k] )
k+=1
if k==26:
j+=1
k=0
i+=1
print "===split SR:==="
splitSR_cmd = "split -l " + str(Nsplitline) + " " + SR_pathfilename + " " + output_path +SR_filename +"."
print splitSR_cmd
log_command(splitSR_cmd)
##########################################
print "===compress SR.aa:==="
i=0
T_blat_SR_ls = []
for ext in ext_ls:
blat_SR_cmd = blat_path + " " + blat_option + ' ' + output_path + SR_filename + ext + ' ' + output_path + SR_filename + ext + ".psl"
print blat_SR_cmd
T_blat_SR_ls.append( threading.Thread(target=log_command, args=(blat_SR_cmd,)) )
T_blat_SR_ls[i].start()
i+=1
for T in T_blat_SR_ls:
T.join()
def main():
# Read input parameters
bin_path,command = GetPathAndName(sys.argv[0])
regions_filename = sys.argv[1]
reads_filename = sys.argv[2]
num_threads = int(sys.argv[3])
python_path = sys.argv[4]
read_len = sys.argv[5]
min_junction_overlap_len = sys.argv[6]
output_filename = 'refSeq_MLE_input.txt'
reads_file = open(reads_filename, 'r' )
reads_files = []
for thread_idx in range(num_threads):
reads_files.append(open(reads_filename + '.' + str(thread_idx), 'w'))
thread_idx = 0
for line in reads_file:
reads_files[thread_idx].write(line)
thread_idx = (thread_idx + 1) % num_threads
for thread_idx in range(num_threads):
reads_files[thread_idx].close()
reads_file.close()
##############################
threads_list = []
for thread_idx in range(num_threads):
cmd = (python_path + " " + bin_path + 'parseSAM.py ' + regions_filename + ' ' + reads_filename + '.' + str(thread_idx) +
' ' + output_filename + '.' + str(thread_idx) + ' ' + read_len + ' ' + min_junction_overlap_len)
print cmd
threads_list.append( threading.Thread(target=log_command, args=(cmd,)) )
threads_list[thread_idx].start()
for thread in threads_list:
thread.join()
output_file = open(output_filename, 'w')
output_files = []
header = 0
for thread_idx in range(num_threads):
output_files.append(open(output_filename + '.' + str(thread_idx), 'r'))
header += int(output_files[thread_idx].readline())
output_file.write(str(header) + '\n')
genes_str_map = {}
genes_reads_count_map = {}
for thread_idx in range(num_threads):
while True:
line = output_files[thread_idx].readline()
if (line == ''):
break
if not genes_str_map.has_key(line):
lines = line
isoforms_line = output_files[thread_idx].readline()
lines += isoforms_line
for i in range(4):
lines += output_files[thread_idx].readline()
for i in range(len(isoforms_line.split())):
lines += output_files[thread_idx].readline()
lines += output_files[thread_idx].readline()
genes_reads_count_map[line] = [int(i) for i in output_files[thread_idx].readline().split()]
genes_str_map[line] = lines
else:
isoforms_line = output_files[thread_idx].readline()
for i in range(4):
output_files[thread_idx].readline()
for i in range(len(isoforms_line.split())):
output_files[thread_idx].readline()
output_files[thread_idx].readline()
reads_count_line = [int(i) for i in output_files[thread_idx].readline().split()]
for i in range(len(reads_count_line)):
genes_reads_count_map[line][i] += reads_count_line[i]
genes = sorted(genes_str_map.keys())
for gene in genes:
output_file.write(genes_str_map[gene])
for i in range(len(genes_reads_count_map[gene])):
output_file.write(str(genes_reads_count_map[gene][i]).ljust(20))
output_file.write('\n')
for thread_idx in range(num_threads):
output_files[thread_idx].close()
rm_cmnd = "rm " + output_filename + '.' + str(thread_idx) + ' ' + reads_filename + '.' + str(thread_idx)
log_command(rm_cmnd)
output_file.close()
def main():
# Read input parameters
bin_path,command = GetPathAndName(sys.argv[0])
input_filename = sys.argv[1]
output_filename = sys.argv[2]
num_threads = int(sys.argv[3])
python_path = sys.argv[4]
penalty_filename = ''
if (len(sys.argv) > 5):
penalty_filename = sys.argv[5]
input_file = open(input_filename, 'r' )
header = input_file.readline()
input_files = []
output_filenames = []
for thread_idx in range(num_threads):
input_files.append(open(input_filename + '.' + str(thread_idx), 'w'))
input_files[-1].write(header)
thread_idx = 0
while True:
line = input_file.readline()
if line == "":
break
num_isoforms = int(line.split()[1])
input_files[thread_idx].write(line)
for i in range(6):
input_files[thread_idx].write(input_file.readline())
for i in range(num_isoforms):
input_files[thread_idx].write(input_file.readline())
for i in range(2):
input_files[thread_idx].write(input_file.readline())
thread_idx = (thread_idx + 1) % num_threads
for thread_idx in range(num_threads):
input_files[thread_idx].close()
input_file.close()
##############################
threads_list = []
for thread_idx in range(num_threads):
cmd = python_path + " " + bin_path + 'MLE_regions.py ' + input_filename + '.' + str(thread_idx) + ' ' + output_filename + '.' + str(thread_idx)
if (penalty_filename != ''):
cmd += ' ' + penalty_filename
print cmd
threads_list.append( threading.Thread(target=log_command, args=(cmd,)) )
threads_list[thread_idx].start()
for thread in threads_list:
thread.join()
cat_cmnd = 'cat '
rm_cmnd = "rm "
for thread_idx in range(num_threads):
cat_cmnd += output_filename + '.' + str(thread_idx) + " "
rm_cmnd += output_filename + '.' + str(thread_idx) + " " + input_filename + '.' + str(thread_idx) + " "
cat_cmnd += ' > ' + output_filename
log_command(cat_cmnd)
log_command(rm_cmnd)
