def multiWellington(regions, reads, **kwargs):
p = mp.Pool(CPUs)
for i in progress.bar(regions):
if clargs.one_dimension:
fp = footprinting.wellington1D(i, reads, **kwargs)
else:
fp = footprinting.wellington(i, reads, **kwargs)
p.apply_async(fp, callback=writetodisk)
# Hold here while the queue is bigger than the number of reads we're happy to store in memory
while p._taskqueue.qsize() > max_regions_cached_in_memory:
pass
p.close()
puts_err("Waiting for the last {0} jobs to finish...".format(max_regions_cached_in_memory))
p.join()
def multiWellington(regions,reads,**kwargs):
p = mp.Pool(CPUs)
for i in progress.bar(regions):
if clargs.one_dimension:
fp = footprinting.wellington1D(i,reads,**kwargs)
else:
fp = footprinting.wellington(i,reads,**kwargs)
p.apply_async(fp,callback = writetodisk)
#Hold here while the queue is bigger than the number of reads we're happy to store in memory
while p._taskqueue.qsize() > max_regions_cached_in_memory:
pass
p.close()
puts_err("Waiting for the last {0} jobs to finish...".format(max_regions_cached_in_memory))
p.join()
#Call footprints
import sys
import pyDNase
import pyDNase.footprinting as fp
if (sys.argv[5] == 'singleEnd'):
regions = pyDNase.GenomicIntervalSet(sys.argv[1])
reads = pyDNase.BAMHandler(sys.argv[2])
f = len(regions) - 1
for x in range(f):
footprinter = fp.wellington1D(regions[x], reads)
footprints = footprinter.footprints(withCutoff=int(sys.argv[4]))
with open(sys.argv[3], "a") as bedout:
bedout.write(str(footprints))
else:
regions = pyDNase.GenomicIntervalSet(sys.argv[1])
reads = pyDNase.BAMHandler(sys.argv[2])
f = len(regions) - 1
for x in range(f):
footprinter = fp.wellington(regions[x], reads)
footprints = footprinter.footprints(withCutoff=int(sys.argv[4]))
with open(sys.argv[3], "a") as bedout:
bedout.write(str(footprints))
#Call footprints import sys import pyDNase import pyDNase.footprinting as fp if(sys.argv[5] == 'singleEnd'): regions = pyDNase.GenomicIntervalSet(sys.argv[1]) reads = pyDNase.BAMHandler(sys.argv[2]) f=len(regions)-1 for x in range(f): footprinter = fp.wellington1D(regions[x],reads) footprints = footprinter.footprints(withCutoff=int(sys.argv[4])) with open(sys.argv[3],"a") as bedout: bedout.write(str(footprints)) else: regions = pyDNase.GenomicIntervalSet(sys.argv[1]) reads = pyDNase.BAMHandler(sys.argv[2]) f=len(regions)-1 for x in range(f): footprinter = fp.wellington(regions[x],reads) footprints = footprinter.footprints(withCutoff=int(sys.argv[4])) with open(sys.argv[3],"a") as bedout: bedout.write(str(footprints))
wigout = open(os.path.join(args.outputdir,args.output_prefix+".wig"),"w")
fdrout = open(os.path.join(args.outputdir,args.output_prefix+".FDR.{0}.bed".format(args.FDR_cutoff)),"w")
#Required for UCSC upload
print >> wigout, "track type=wiggle_0"
#Iterate in chromosome, basepair order
orderedbychr = [item for sublist in sorted(regions.intervals.values())
for item in sorted(sublist, key=lambda peak: peak.startbp)]
#puts("Calculating footprints...")
#for each in progress.bar(orderedbychr):
#Calculate footprint scores (1D or 2D)
#TODO: put args here.
for each in orderedbychr:
if args.one_dimension:
fp = footprinting.wellington1D(each, reads, shoulder_sizes=args.shoulder_sizes,
footprint_sizes=args.footprint_sizes, bonferroni=args.bonferroni)
else:
fp = footprinting.wellington(each, reads, shoulder_sizes=args.shoulder_sizes,
footprint_sizes=args.footprint_sizes, bonferroni=args.bonferroni)
#Write fpscores to WIG
print >> wigout, "fixedStep\tchrom="+str(fp.interval.chromosome)+"\tstart="+ str(fp.interval.startbp)+"\tstep=1"
for i in fp.scores:
print >> wigout, i
#FDR footprints
fdr = percentile(np.concatenate([fp.calculate(reads, FDR=True, shoulder_sizes=args.shoulder_sizes,
footprint_sizes=args.footprint_sizes, bonferroni=args.bonferroni)[0]
for i in range(args.FDR_iterations)]).tolist(),args.FDR_cutoff)
### print fdr
if fdr < args.FDR_limit: