def main(argv=None):
P.main(argv)
bandwidth = PARAMS["plot_bandwidth"]
script = PARAMS["pipeline_dir"] + "R/plotFootprint.R"
statement = '''Rscript {script}
--infiles {infiles}
--outfile {outfile}
--labels {labels}
--title {region}
-b {bandwidth}
--xlims {distance}
--unsmoothed {smoothed}'''
P.run(statement)
# ---------------------------------------------------
# Generic pipeline tasks
@follows(plotFootprint, plotFootprints)
def full():
pass
def main(argv=None):
if argv is None:
argv = sys.argv
P.main(argv)
if __name__ == "__main__":
sys.exit(P.main(sys.argv))
def count_genes(infile, outfile):
statement = '''wc -l %(infile)s > %(outfile)s''' #counts no. of lines, s specifies string. Each line = transcript
P.run(statement) #if on cbrg have to specify job queue. Will run and save to single file per chromosome
#or can do len(chr1.gtf) - count number of lines in file
@merge(count_genes,'all.average')
def average (infiles, outfile): #each count file has 2 items e.g 100 (no. of transcripts) chr1.gtf (original file name)
total_counts = {} #create dictionary
for infile in infiles:
with open (infile) as inf:
count, chrom = inf.read().strip().split(' ') #.read = reading the line, .strip = taking away white space, .split = setting 2 variables
total_counts[chrom] = int(count) #key = chrom, count (make integer) = value
median = statistics.median(total_counts.values()) #calculate median from integers only
with open(outfile,'w') as count:
for key, value in total_counts.items():
count.write(f'{key}\t{value}\n') #write dictionary
count.write(f'Median\t{median}\n') #write median
if __name__ == "__main__":
sys.exit( P.main(sys.argv) ) #if have this at bottom can run in command line eg $python workflow.py make for full pipeline
#If want to set input file using .yml then make .yml file with (save .yml file in same directory):
gtf = /ifs/obds-training/lingf/obds/devel/OBDS_Training_Sep_2019/genes.gtf.gz
#Then under import statements
P.get_parameters('workflow.yml')
#And have to change decorators e.g
@split(P.PARAMS['gtf'], 'chr*.gtf') #end up with dictionary, gtf is key and value is file path
def main(argv=None):
if argv is None:
argv = sys.argv
P.main(argv)
def main(argv=None):
if argv is None:
argv = sys.argv
P.main(argv)
if __name__ == "__main__":
sys.exit(P.main(sys.argv))
def main():
P.main(sys.argv)
token = token[0]
else:
token = None
s, infile = Sra.process_remote_BAM(infile,
token,
tmpfilename,
filter_bed=PARAMS["contigs_bed"])
s = re.sub(";\n", " &&\n", s)
infile = ",".join(infile)
statement = " && ".join([
"mkdir -p %(tmpfilename)s", s, statement, "rm -r %(tmpfilename)s"
])
P.run(statement,
job_condaenv=PARAMS["rmats_env"],
job_memory=PARAMS["rmats_prep_memory"])
rmats_counter = ""
for f_path in glob("rmats.dir/%(track)s.dir/*.rmats" % locals()):
shutil.copy(f_path,
P.snip(outfile, ".rmats") + rmats_counter + ".rmats")
if rmats_counter == "":
rmats_counter = 1
else:
rmats_counter += 1
P.main(sys.argv)
genomegenes.add(gene)
if g in genomegenes:
result[g].append("1")
else:
result[g].append("0")
outf.write("\t".join(
["gene"] + [P.snip(os.path.basename(x), ".tsv")
for x in infiles]) + "\n")
for gene, data in result.items():
outf.write("\t".join([gene] + data) + "\n")
outf.close()
##############################################
##############################################
##############################################
@follows(mergeAnnotations)
def full():
pass
##############################################
##############################################
##############################################
if __name__ == "__main__":
P.main()
def main():
sys.exit(P.main(sys.argv))
