def read_scores(f):
scores_by_chrom = dict()
for chrom, pos, val in bx.wiggle.Reader(f):
if chrom not in scores_by_chrom:
scores_by_chrom[chrom] = BinnedArray()
scores_by_chrom[chrom][pos] = val
return scores_by_chrom
def load_scores_wiggle(fname, chrom_buffer_size=3):
"""
Read a wiggle file and return a dict of BinnedArray objects keyed
by chromosome.
"""
scores_by_chrom = dict()
try:
for chrom, pos, val in bx.wiggle.Reader(UCSCOutWrapper(open(fname))):
if chrom not in scores_by_chrom:
if chrom_buffer_size:
scores_by_chrom[chrom] = BinnedArray()
chrom_buffer_size -= 1
else:
scores_by_chrom[chrom] = PositionalScoresOnDisk()
scores_by_chrom[chrom][pos] = val
except UCSCLimitException:
# Wiggle data was truncated, at the very least need to warn the user.
print(
'Encountered message from UCSC: "Reached output limit of 100000 data values", so be aware your data was truncated.'
)
except IndexError:
stop_err(
'Data error: one or more column data values is missing in "%s"' %
fname)
except ValueError:
stop_err(
'Data error: invalid data type for one or more values in "%s".' %
fname)
return scores_by_chrom
def load_scores_wiggle( fname ):
"""
Read a wiggle file and return a dict of BinnedArray objects keyed
by chromosome.
"""
scores_by_chrom = dict()
for chrom, pos, val in bx.wiggle.Reader( misc.open_compressed( fname ) ):
if chrom not in scores_by_chrom:
scores_by_chrom[chrom] = BinnedArray()
scores_by_chrom[chrom][pos] = val
return scores_by_chrom
def main():
# Parse command line
options, args = doc_optparse.parse( __doc__ )
try:
if options.comp:
comp_type = options.comp
else:
comp_type = None
score_fname = args[0]
out_fname = args[1]
except:
doc_optparse.exit()
scores = BinnedArray()
## last_chrom = None
for i, ( chrom, pos, val ) in enumerate( bx.wiggle.Reader( misc.open_compressed( score_fname ) ) ):
#if last_chrom is None:
# last_chrom = chrom
#else:
# assert chrom == last_chrom, "This script expects a 'wiggle' input on only one chromosome"
scores[pos] = val
# Status
if i % 10000 == 0: print i, "scores processed"
out = open( out_fname, "w" )
if comp_type:
scores.to_file( out, comp_type=comp_type )
else:
scores.to_file( out )
out.close()
def __init__(self, wigFile):
'''read wig file, creat wig obj'''
self.scores = {}
self.num_re = re.compile(r'[\d\.\-\+]+')
fh = open(wigFile)
#infile=open(wigFile,'r')
for i, (chrom, pos, val) in enumerate(bx.wiggle.Reader(fh)):
chrom = chrom.upper()
if not chrom in self.scores: self.scores[chrom] = BinnedArray()
self.scores[chrom][pos] = val
if i % 100000 == 0: print "%i datapoints loaded \r" % i
#print self.scores.keys()
print "total " + str(i) + " points loaded"
def main():
options, args = doc_optparse.parse(__doc__)
scores = {}
for i, (chrom, pos, val) in enumerate(bx.wiggle.Reader(open(sys.argv[1]))):
if chrom not in scores:
scores[chrom] = BinnedArray()
scores[chrom][pos] = val
# Status
if i % 10000 == 0:
print(i, "scores processed")
for chr in scores.keys():
out = open(chr, "w")
scores[chr].to_file(out)
out.close()
def read_WIG(wig_file):
wig = open(wig_file, "r")
ribo_cov = {}
for line in wig:
if line.startswith('variableStep'): # specific for Shoelaces wig files
stepType = 'variable'
fields = line.split()[1:]
declarations = dict([(p[0], p[1].strip('"')) for p in [x.split("=") for x in fields]])
chrom = declarations['chrom']
span = 1
ribo_cov[chrom] = BinnedArray()
else:
tmp = line.strip().split()
pos = int(tmp[0])
val = float(tmp[1])
ribo_cov[chrom][pos] = val
wig.close()
return ribo_cov
def main():
# Parse command line
options, args = doc_optparse.parse( __doc__ )
try:
score_fname = args[0]
except:
doc_optparse.exit()
scores = {}
for i, ( chrom, pos, val ) in enumerate( bx.wiggle.Reader( open(sys.argv[1]) ) ):
if not chrom in scores: scores[ chrom ] = BinnedArray()
scores[chrom][pos] = val
# Status
if i % 10000 == 0: print i, "scores processed"
for chr in scores.keys():
out = open( chr, "w" )
scores[chr].to_file( out )
out.close()
#!/usr/bin/env python
from bx.binned_array import BinnedArray
import os
COMPTYPE = 'zlib'
TYPECODE = 'I'
newarray = lambda: BinnedArray(default=0, typecode=TYPECODE)
def open_with_newdir(filename):
dname = os.path.dirname(filename)
if not os.path.isdir(dname):
os.makedirs(dname)
return open(filename, 'w')
BASESPACE = ['depth', 'A', 'C', 'G', 'T', 'N']
BASESPACE2i = dict((n, i) for i, n in enumerate(BASESPACE))
def process(inpf, prefix):
arrchrom = None
arrtype = None
arr = []
def flush():
if arrchrom is None:
return
filenamefmt = '%s/%%s/%s.%s' % (prefix, arrchrom[0], arrchrom[1])
if arrtype != 'M':
