def fa_cuts(fname,e1,e2):
listfile = '%s_%s-%s.list' % (fname,e1,e2)
if os.path.exists(listfile):
print >> sys.stderr, '%s found, load fragment lengths ...' % listfile,
fraglens = eval(open(listfile).read())
print >> sys.stderr, '%s fragments loaded' % len(fraglens)
return fraglens
fraglens = []
this = []
seqnum = 0
for l in open(fname):
if l[0] == '>':
if this:
seq = ''.join(this)
fraglens.extend(fraglen_from_seq(seq,e1,e2))
this = []
sname = l[1:].strip()
seqnum += 1
outn = ' (sequence %s)' % (seqnum)
om = '\r%s %s' % (sname[:80-len(outn)], outn)
print >> sys.stderr, om, ' '*(80-len(om)+1),
else:
this.append(l.strip().upper())
seq = ''.join(this)
fraglens.extend(fraglen_from_seq(seq,e1,e2))
print >> sys.stderr, 'store %s fragment lengths to %s' % (len(fraglens),listfile)
open(listfile,'w').write(fraglens.__repr__())
return fraglens
def fa_cuts(fname, e1, e2):
listfile = '%s_%s-%s.list' % (fname, e1, e2)
if os.path.exists(listfile):
print >> sys.stderr, '%s found, load fragment lengths ...' % listfile,
fraglens = eval(open(listfile).read())
print >> sys.stderr, '%s fragments loaded' % len(fraglens)
return fraglens
fraglens = []
this = []
seqnum = 0
for l in open(fname):
if l[0] == '>':
if this:
seq = ''.join(this)
fraglens.extend(fraglen_from_seq(seq, e1, e2))
this = []
sname = l[1:].strip()
seqnum += 1
outn = ' (sequence %s)' % (seqnum)
om = '\r%s %s' % (sname[:80 - len(outn)], outn)
print >> sys.stderr, om, ' ' * (80 - len(om) + 1),
else:
this.append(l.strip().upper())
seq = ''.join(this)
fraglens.extend(fraglen_from_seq(seq, e1, e2))
print >> sys.stderr, 'store %s fragment lengths to %s' % (len(fraglens),
listfile)
open(listfile, 'w').write(fraglens.__repr__())
return fraglens
def load_cluster_data(gr,tab,mID=None):
'''given an mcl graph output file, an mcl tab file of sequence labels, and an mID file of individuals per label,
returns (clusters,labels,mID_by_label) dicts
'''
#extract clusters from mcl output
print >>sys.stderr, 'load graph...',
fc = open(gr).read()
print >> sys.stderr, 'done\nparse graph...',
body = re.search('begin(.+?)\)',fc.replace('\n',' ')).groups()[0]
clusters = dict([(s.strip().split()[0],s.strip().split()[1:]) for s in body.strip().split('$') if len(s.strip().split()) > 1])
print >> sys.stderr, 'done\nload labels...',
#cluster labels
labels = dict([l.strip().split() for l in open(tab).readlines()])
print >> sys.stderr, 'done\nload mIDs...',
#individuals by labels
if mID is None:
mID_by_label = None
else:
mID_by_label = dict([(l.split()[0],l.split()[1:]) for l in open(mID).readlines()])
print >> sys.stderr, 'done'
return clusters,labels,mID_by_label
def load_cluster_data(gr, tab, mID=None):
'''given an mcl graph output file, an mcl tab file of sequence labels, and an mID file of individuals per label,
returns (clusters,labels,mID_by_label) dicts
'''
#extract clusters from mcl output
print >> sys.stderr, 'load graph...',
fc = open(gr).read()
print >> sys.stderr, 'done\nparse graph...',
body = re.search('begin(.+?)\)', fc.replace('\n', ' ')).groups()[0]
clusters = dict([(s.strip().split()[0], s.strip().split()[1:])
for s in body.strip().split('$')
if len(s.strip().split()) > 1])
print >> sys.stderr, 'done\nload labels...',
#cluster labels
labels = dict([l.strip().split() for l in open(tab).readlines()])
print >> sys.stderr, 'done\nload mIDs...',
#individuals by labels
if mID is None:
mID_by_label = None
else:
mID_by_label = dict([(l.split()[0], l.split()[1:])
for l in open(mID).readlines()])
print >> sys.stderr, 'done'
return clusters, labels, mID_by_label
def load_lines_from_uniqued(source_uniques,
rv_sort=True,
sort_key=lambda x: (len(x[0]), int(x[1])),
keep_source_id=False):
'''
if keep_source_id is True
returns list of 2-tuples uniqued_id (eg 100617_lane6_PE for "data/100617/100617_lane6_PE.uniqued")
tuples are (parsed_lines,uniqued_id)
else list of lines.
'''
uniquedlines = []
for f in source_uniques:
lines = []
print >> sys.stderr, 'load %s ...' % f,
lines = tuple([l.strip().split() for l in open(f).readlines()])
print >> sys.stderr, '%s lines' % len(lines)
#get qual base
baseQ = None
for l in lines:
baseQ = get_baseQ(l[2])
if baseQ is not None:
break
print >> sys.stderr, 'qual base: %s' % baseQ
if baseQ == 64:
print >> sys.stderr, 'Translate quality encoding to base 33 ...',
for l in lines:
l[2] = ''.join([chr(ord(c) - 64 + 33) for c in l[2]])
print >> sys.stderr, 'done'
if keep_source_id:
uniqued_id = os.path.basename(os.path.splitext(f)[0])
uniquedlines.extend(zip(lines, [uniqued_id] * len(lines)))
else:
uniquedlines.extend(lines)
print >> sys.stderr, 'sort',
if keep_source_id:
uniquedlines.sort(reverse=rv_sort, key=lambda x: sort_key(x[0]))
else:
uniquedlines.sort(reverse=rv_sort, key=sort_key)
print >> sys.stderr, 'done'
return uniquedlines
def load_lines_from_uniqued(source_uniques,rv_sort = True, sort_key = lambda x: (len(x[0]),int(x[1])), keep_source_id = False):
'''
if keep_source_id is True
returns list of 2-tuples uniqued_id (eg 100617_lane6_PE for "data/100617/100617_lane6_PE.uniqued")
tuples are (parsed_lines,uniqued_id)
else list of lines.
'''
uniquedlines = []
for f in source_uniques:
lines = []
print >> sys.stderr, 'load %s ...' % f,
lines = tuple([l.strip().split() for l in open(f).readlines()])
print >> sys.stderr, '%s lines' % len(lines)
#get qual base
baseQ = None
for l in lines:
baseQ = get_baseQ(l[2])
if baseQ is not None:
break
print >> sys.stderr, 'qual base: %s' % baseQ
if baseQ == 64:
print >> sys.stderr, 'Translate quality encoding to base 33 ...',
for l in lines:
l[2] = ''.join([chr(ord(c)-64+33) for c in l[2]])
print >> sys.stderr, 'done'
if keep_source_id:
uniqued_id = os.path.basename(os.path.splitext(f)[0])
uniquedlines.extend( zip( lines,[uniqued_id]*len(lines) ) )
else:
uniquedlines.extend(lines)
print >> sys.stderr, 'sort',
if keep_source_id:
uniquedlines.sort(reverse = rv_sort,key = lambda x: sort_key(x[0]))
else:
uniquedlines.sort(reverse = rv_sort,key = sort_key)
print >> sys.stderr, 'done'
return uniquedlines
