def main(parclipA, parclipB, start, stop, width, anno=None, annowidth=100,
logRatio=False, verbose=False):
tmpA = ParclipSiteContainer()
dataB = ParclipSiteContainer()
tmpA.loadFromFile(parclipA)
tmpA.sort(key='occ')
dataB.loadFromFile(parclipB)
if start < 0 or stop < start or stop >= tmpA.size():
print('Bullshit start and stop indices. Come on! Concentrate!')
sys.exit()
dataA = parclipsites.ParclipSites('')
total = stop - start
count = 0
i = start
while count < total and i < (tmpA.size()-1):
if verbose:
functions.showProgress(count,total-1,'Selecting PAR-CLIP sites')
if anno == None:
dataA.addSite(tmpA.chrs[i], tmpA.pos[i], tmpA.m[i], tmpA.r[i],
tmpA.result[i], tmpA.strand[i], tmpA.occ[i])
count +=1
else:
if anno.isInside(tmpA.chrs[i], tmpA.pos[i], tmpA.strand[i],
annowidth, annowidth)[1]:
dataA.addSite(tmpA.chrs[i], tmpA.pos[i], tmpA.m[i], tmpA.r[i],
tmpA.result[i], tmpA.strand[i], tmpA.occ[i])
count +=1
i += 1
coloc = 1
count_coloc = 1
if verbose:
print('\n')
for i in range(dataA.size()):
values = dataB.getValues(dataA.chrs[i], dataA.pos[i], dataA.strand[i], True, width, width)
if values != None:
count_coloc += 1
coloc += max(values)
if verbose:
functions.showProgress(i, (dataA.size()-1), 'Collecting colocolization data')
coloc = coloc / count_coloc
if verbose:
print('')
if logRatio:
return math.log( coloc/functions.getQuantile(dataB.occ,0.5) ,2)
else:
return coloc
from mockinbird.utils import ParclipSiteContainer
if __name__ == '__main__':
parser = argparse.ArgumentParser(description='Takes PAR-CLIP sites and a genome and saves genomic sequences as fasta file around PAR-CLIP sites according to the given parameters.', epilog="contact: [email protected]")
parser.add_argument('sites', help='PAR-CLIP file *.table')
parser.add_argument('genome', help='path to genome')
parser.add_argument('fafile', help='output filename')
parser.add_argument('filterGFF', help='set path to GFF if sites should be removed that overlap with the GFF [default = '']', default='')
parser.add_argument('start', help='start index of PAR-CLIP sites [default=0]', type=int, default = 0)
parser.add_argument('stop', help='stop index of PAR-CLIP sites [default=1500]', type=int, default = 1500)
parser.add_argument('width', help='number of nt +/- the crosslink site [default=15]', type=int, default = 15)
parser.add_argument('additionalFilterWidth', help='number of nt that are added to the start/stop indices of the GFF annotations', type=int, default = 20)
parser.add_argument('key', help='set key that is used for PAR-CLIP site ordering [default = \'occ\'], options: [\'occ\', \'m\', \'r\', \'mr\', \'pvalue\']', default='occ')
parser.add_argument('-v','--verbose', dest='verbose', action="store_true", default=False, help='verbose output')
args = parser.parse_args()
yeast = genome.Genome(args.genome, False)
sites = ParclipSiteContainer()
sites.loadFromFile(args.sites)
if args.verbose:
print('#sites : '+str(sites.size()))
if args.filterGFF != '':
anno = gff.GFF(args.filterGFF)
sites = sites.removeSitesLocatedInGFF(anno, args.additionalFilterWidth)
print('#sites after removal: '+str(sites.size()))
sites.sort(args.key)
sites.save2Fasta(yeast, args.fafile, args.start, args.stop, args.width)