def saveNexus0(self, oFileHandle, datatype='protein', gap='-', interleave=False, width=60):
assert datatype.lower() in ['dna', 'rna', 'protein', 'standard', 'restriction']
yesno = {True: 'yes', False: 'no'}
oFile = smartopen(oFileHandle, 'w')
print >> oFile, '#NEXUS'
print >> oFile, 'BEGIN data;'
print >> oFile, 'DIMENSIONS ntax=%i nchar=%i;' % (self.numberOfSeqs(), self.__len__())
print >> oFile, 'FORMAT datatype=%s interleave=%s gap=%s;' \
% (datatype, yesno[interleave], gap)
print >> oFile
print >> oFile, 'MATRIX'
format = '%%-%is %%s' % max([len(name) for name in self.seqDict])
if interleave:
for i in xrange(0,self.__len__(),width):
for name in self.order:
print >> oFile, format % (name, self.seqDict[name][i:i+width])
print >> oFile
else:
for name in self.order:
print >> oFile, format % (name, self.seqDict[name])
print >> oFile
print >> oFile, ';'
print >> oFile, 'END;'
oFile.close()
def loadPhylip(iFileHandle, multipleDatasets=False):
"""
Load phylip alignment data.
@param iFileHandle: Input filename or file.
@param multipleDatasets: Do not close file (default False)
"""
iFile = smartopen(iFileHandle)
nSeq,L = [int(x) for x in iFile.readline().strip().split()]
alignment = Alignment()
for i in xrange(nSeq):
line = iFile.readline()
tokens = line.strip().split()
seq = ''.join(tokens[1:])
alignment.append(tokens[0], seq)
skip = iFile.readline()
width = len(alignment)
nBlocks = int(math.ceil(float(L)/width))
for j in xrange(nBlocks-1):
for i in xrange(nSeq):
line = iFile.readline().strip()
name = alignment.getName(i)
alignment.append(name, line, cleanup=True)
try:
skip = iFile.readline()
except:
pass
if not multipleDatasets: iFile.close()
return alignment
def load(iFilename, offset=0):
"""Load a BED file.
@param iFilename: Input filename or file.
@param offset: Offset subtracted from positions (Default: 0).
@return: List of features.
"""
iFile = smartopen(iFilename)
data = []
for line in iFile:
line = line.strip()
if line and line[0] != "#":
tokens = line.split("\t")
f = Feature(tokens)
try:
f.chromStart -= offset
f.chromEnd -= offset
f.thickStart -= offset
f.thickEnd -= offset
except:
pass
data.append(f)
return data
def saveMolphy(self, oFileHandle, width=60):
oFile = smartopen(oFileHandle, 'w')
print >> oFile, '%i %i' % (self.numberOfSeqs(), self.__len__())
for name in self.order:
print >> oFile, name
for i in xrange(0, self.__len__(), width):
print >> oFile, self.seqDict[name][i:i+width]
oFile.close()
def load2(iFileHandle):
iFile = smartopen(iFileHandle)
genes = []
for line in iFile:
tokens = line.strip().split()
if not line or line[0]=='#':
continue
elif line[0] in ['<', '>']:
genes.append(Gene.fromTokens(tokens))
else:
genes[-1].add(Exon.fromTokens(tokens))
def __init__(self, fileHandle, mode='w', width=60, blockSize=None, **kw):
"""
@param iFileHandle: Output file or name
@keyword mode: File mode - write(w) or append(a)
"""
assert mode in ('w', 'a')
self.iFile = smartopen(fileHandle, mode)
self.iFilename = self.iFile.name
if kw:
print 'Uncaptured keywords'
print kw
def load(iFileHandle):
"""Load WIG file.
@param iFileHandle: Input file or filename
@return: (header, data)
"""
iFile = smartopen(iFileHandle)
header = iFile.readline()
data = []
for line in iFile:
score = float(line.strip())
data.append(score)
return header,data
def load_iter(iFileHandle, skip=5, splitOn=None):
"""Load TRANSFAC match output.
Arguments:
iFileHandle -- Input file or filename.
"""
iFile = smartopen(iFileHandle)
for i in xrange(skip):
iFile.next()
for line in iFile:
tokens = line.strip().split(splitOn)
yield Match(tokens)
def load_iter(iFileHandle):
iFile = smartopen(iFileHandle)
first = True
for line in iFile:
tokens = line.strip().split()
if not line or line[0]=='#':
continue
elif line[0] in ['<', '>']:
if not first:
yield gene
else:
first = False
gene = Gene.fromTokens(tokens)
else:
gene.add(Exon.fromTokens(tokens))
yield gene
def savePhylip(self, oFileHandle, width=10, **kw):
oFile = smartopen(oFileHandle, 'w')
print >> oFile, '%7i%7i' % (self.numberOfSeqs(), self.__len__())
L = self.__len__()
format = '%%-%is %%s' % width
for i in xrange(0,L,50):
for name in self.order:
if i==0:
label = name[0:width]
else:
label = ''
print >> oFile, ('%%-%is' % width) % label,
for j in xrange(0,50,10):
print >> oFile, self.seqDict[name][i+j:i+j+10],
print >> oFile
print >> oFile
oFile.close()
def __init__(self, iFileHandle, force=False, section=None, **kw):
"""Constructor
@param iFileHandle: CAP3 file name or object
"""
self.iFile = smartopen(iFileHandle)
self.iFilename = self.iFile.name
self.indexFile = CapIndexFile(self.iFilename)
self.indexFile.build(force=force)
self.stopAtMiddle = False
if section==1:
self.stopAtMiddle = True
elif section==2:
self.seek(0,2)
self._iter = None
self._initIter = True
self._section = section
def load_iter(iFileHandle, format='psl', **kw):
"""Return an iterator to the BLAT file.
@param iFileHandle: Input filename or file.
@param format: BLAT output format (optional; default: 'psl')
"""
if not format in ['psl']:
raise 'Only psl is currently supported.'
iFile = smartopen(iFileHandle)
skip = kw.pop('skip', 5)
for i in xrange(skip):
junk = iFile.readline()
for line in iFile:
if line:
tokens = line.strip().split('\t')
yield Chain(tokens, **kw)
def saveNexus(self, oFileHandle, datatype='protein', gap='-', interleave=False, width=60):
assert datatype.lower() in ['dna', 'rna', 'protein', 'standard', 'restriction']
yesno = {True: 'yes', False: 'no'}
oFile = smartopen(oFileHandle, 'w')
print >> oFile, '#nexus'
print >> oFile
print >> oFile, 'BEGIN Taxa;'
print >> oFile, 'DIMENSIONS ntax=%i;' % self.numberOfSeqs()
print >> oFile, 'TAXLABELS'
for i,name in enumerate(self.order):
print >> oFile, "[%i] '%s'" % (i+1,name)
print >> oFile, ';'
print >> oFile, 'END; [Taxa]'
print >> oFile
print >> oFile, 'BEGIN Characters;'
print >> oFile, 'DIMENSIONS nchar=%i;' % self.__len__()
print >> oFile, 'FORMAT'
print >> oFile, ' datatype=%s' % datatype
print >> oFile, ' missing=?'
print >> oFile, ' gap=%s' % gap
print >> oFile, ' symbols="a r n d c q e g h i l k m f p s t w y v z"'
print >> oFile, ' labels=left'
print >> oFile, ' transpose=no'
print >> oFile, ' interleave=%s' % yesno[interleave]
print >> oFile, ';'
print >> oFile, 'MATRIX'
format = "%%-%is %%s" % max([len(name) for name in self.seqDict])
if interleave:
for i in xrange(0,self.__len__(),width):
for name in self.order:
print >> oFile, format % (name, self.seqDict[name][i:i+width])
print >> oFile
else:
for name in self.order:
print >> oFile, format % (name, self.seqDict[name])
print >> oFile
print >> oFile, ';'
print >> oFile, 'END;'
oFile.close()
def loadClustal(iFileHandle, headerCheck=True):
"""
Load clustal alignment data.
@param iFileHandle: Input filename or file.
@param headerCheck: Test that CLUSTAL appears on first line (default True)
"""
iFile = smartopen(iFileHandle)
clustalHeader = iFile.readline().strip()
if headerCheck and not clustalHeader.split()[0] in ['CLUSTAL', 'MUSCLE']:
raise 'Not a CLUSTAL file'
alignment = Alignment()
for line in iFile:
if line[0]!=' ':
tokens = line.strip().split()
if len(tokens)==2:
alignment.append(tokens[0], tokens[1])
iFile.close()
return alignment
def saveClustal(self, oFileHandle, nameWidth=None, width=60, interleaved=True, **kw):
oFile = smartopen(oFileHandle, 'w')
print >> oFile, 'CLUSTAL W (1.83) multiple sequence alignment\n\n'
if not nameWidth:
nameWidth = max([len(name) for name in self.order])
format = '%%-%is %%s' % nameWidth
L = self.__len__()
if interleaved:
for i in xrange(0,L,width):
for name in self.order:
print >> oFile, format % (name[0:nameWidth], self.seqDict[name][i:i+width])
print >> oFile, format % (' '*nameWidth, ' '*width)
print >> oFile
else:
for name in self.order:
print >> oFile, format % (name[0:nameWidth], self.seqDict[name])
print >> oFile, format % (' '*nameWidth, ' '*len(self.seqDict[name]))
print >> oFile
oFile.close()
def __init__(self, iFileHandle, clobber=False,
interface=Interface.CONTAINER, method=IndexMethod.SQLITE, **kw):
"""
@param iFileHandle: Fasta file name or object
"""
self.iFile = smartopen(iFileHandle)
self.iFilename = self.iFile.name
if method==IndexMethod.PICKLE:
self.indexFile = FastaIndexPickleFile(self.iFilename)
elif method==IndexMethod.TEXT:
self.indexFile = FastaIndexTextFile(self.iFilename)
else: # sqlite3 method is the default
self.indexFile = FastaIndexFile(self.iFilename)
self.indexFile.build(clobber=clobber)
self.interface = interface
self._iter = None
self._initIter = True
if kw:
print 'Uncaptured keywords'
print kw
def loadStockholm(iFileHandle, **kw):
"""Load Stockholm alignment data.
@param iFileHandle: Input filename or file.
@returns: a dictionary of sequences {name1: seq1, name2: seq2, ...}
"""
iFile = smartopen(iFileHandle)
alignment = Alignment()
alignment.headers = {}
# Skip hmmalign header
start = False
for line in iFile:
line = line.strip()
if line == '# STOCKHOLM 1.0':
break
# Parse sto header info
for line in iFile:
line = line.strip()
if not line: # Blank lines
continue
elif line[0:4] == '#=GS': # Fasta headers
header = line.strip()[1:]
tokens = header.split()
alignment.headers[tokens[0]] = header
continue
elif line[0] == '#': # Other boring comment lines
continue
elif line == '//': # End of file
break
# The real stuff
name, seq = line.split()
alignment.append(name, seq)
return alignment
def load_preprocessed(iFileHandle):
"""Load genscan predictions when predictions have been preprocessed
and only contain the gene prediction lines
Arguments:
iFileHandle -- Input file or filename.
Return values:
data -- Annotation data (a list of lists, each list in one gene)
"""
iFile = smartopen(iFileHandle)
data = {}
skipState = 'Slice no. '
state = None
for line in iFile:
line = line.strip()
if line:
if skipState in line:
pass
else:
tokens = line.split()
d = Predicted(tokens)
gene = int(d.gene_exon.split('.')[0])
try:
data[gene].append(d)
except KeyError:
data[gene] = [d]
data = data.items()
data.sort()
data = [x[1] for x in data]
return data
def loadStockholm(iFileHandle, **kw):
"""Load Stockholm alignment data.
@param iFileHandle: Input filename or file.
@returns: a dictionary of sequences {name1: seq1, name2: seq2, ...}
"""
iFile = smartopen(iFileHandle)
alignment = Alignment()
alignment.headers = {}
# Skip hmmalign header
start = False
for line in iFile:
line = line.strip()
if line=='# STOCKHOLM 1.0':
break
# Parse sto header info
for line in iFile:
line = line.strip()
if not line: # Blank lines
continue
elif line[0:4]=='#=GS': # Fasta headers
header = line.strip()[1:]
tokens = header.split()
alignment.headers[tokens[0]] = header
continue
elif line[0]=='#': # Other boring comment lines
continue
elif line=='//': # End of file
break
# The real stuff
name,seq = line.split()
alignment.append(name, seq)
return alignment
def __init__(self, iFileHandle1, iFileHandle2):
self.iFile1 = smartopen(iFileHandle1)
self.iFile2 = smartopen(iFileHandle2)
def load_full(iFileHandle):
"""Load genscan predictions.
Arguments:
iFileHandle -- Input file or filename.
Return values:
data -- Annotation data (a list of lists, each list in one gene)
proteins -- Predicted proteins (a list of tuples (header, sequence))
meta -- Meta-data in first 8 lines of genscan output
"""
iFile = smartopen(iFileHandle)
data = {}
proteins = []
meta = []
startPredState = '----- ---- - ------ ------ ---- -- -- ---- ---- ----- ----- ------'
endPredState = 'Predicted peptide sequence(s):'
skipState = 'Slice no. '
metaState = 'GENSCAN 1.0'
state = None
for line in iFile:
line = line.strip()
if metaState in line:
state = 'meta'
if line == startPredState:
state = 'pred'
elif line == 'NO EXONS/GENES PREDICTED IN SEQUENCE':
state = 'fail'
elif line == endPredState:
state = 'prot'
elif skipState in line:
state = 'skip'
else:
if state == 'meta':
if line:
meta.append(line)
elif state == 'pred':
if line:
tokens = line.split()
d = Predicted(tokens)
gene = int(d.gene_exon.split('.')[0])
try:
data[gene].append(d)
except KeyError:
data[gene] = [d]
elif state == 'prot':
break
elif state == 'fail':
return [], [], ''
if state == 'prot':
proteins = fasta.load_mfa(iFile)
data = data.items()
data.sort()
data = [x[1] for x in data]
return data, proteins, meta
def load(iFileHandle):
iFile = smartopen(iFileHandle)
aln = Alignment()
state = None
for line in iFile:
line = line.rstrip()
if not line:
continue
elif line == '#:lav': # Section break
state = 'BLOCK'
block = Block()
continue
elif line == '#:eof': # End of file
state = 'EOF'
break
elif line[0] == 'd': # Substitution matrix stanza
state = 'MATRIX'
continue
elif line[0] == 's': # Sequence files stanza
state = 'FILES'
continue
elif line[0] == 'h': # Fasta headers stanza
state = 'HEADERS'
continue
elif line[0] == 'a': # Alignment stanza
state = 'ALIGN'
chain = Chain.fromBlock(copy.copy(block))
continue
elif line[0] in ['x', 'm']:
state = 'BORING'
continue
elif state == 'MATRIX' and line[0] == '}':
state = 'MATRIX_END'
aln.matrix = '\n'.join(aln.matrix)
continue
elif state == 'ALIGN' and line[0] == '}':
state = 'ALIGN_END'
aln.chains.append(chain)
chain = None
continue
elif line[0] == '}': # End of state
continue
tokens = line.lstrip().split()
if state == None:
print line
raise Exception('Wrong')
elif state == 'MATRIX':
aln.matrix.append(line)
elif state == 'FILES':
block.filenames.append(tokens[0])
block.lengths.append(int(tokens[2]))
block.strands.append(strandDict[tokens[3]])
# Next line in stanza
tokens = iFile.next().strip().split()
block.filenames.append(tokens[0])
block.lengths.append(int(tokens[2]))
block.strands.append(strandDict[tokens[3]])
elif state == 'HEADERS':
block.headers.append(tokens[0])
# Next line in stanza
tokens = iFile.next().strip().split()
block.headers.append(tokens[0])
elif state == 'ALIGN':
if tokens[0] == 's':
chain.score = int(tokens[1])
elif tokens[0] == 'b':
chain.interval1[0] = int(tokens[1])
chain.interval2[0] = int(tokens[2])
elif tokens[0] == 'e':
chain.interval1[1] = int(tokens[1])
chain.interval2[1] = int(tokens[2])
elif tokens[0] == 'l':
chain.hsps.append(HSP(tokens[1:]))
return aln
def load(iFileHandle):
iFile = smartopen(iFileHandle)
aln = Alignment()
state = None
for line in iFile:
line = line.rstrip()
if not line:
continue
elif line=='#:lav': # Section break
state = 'BLOCK'
block = Block()
continue
elif line=='#:eof': # End of file
state = 'EOF'
break
elif line[0]=='d': # Substitution matrix stanza
state = 'MATRIX'
continue
elif line[0]=='s': # Sequence files stanza
state = 'FILES'
continue
elif line[0]=='h': # Fasta headers stanza
state = 'HEADERS'
continue
elif line[0]=='a': # Alignment stanza
state = 'ALIGN'
chain = Chain.fromBlock(copy.copy(block))
continue
elif line[0] in ['x', 'm']:
state = 'BORING'
continue
elif state=='MATRIX' and line[0]=='}':
state = 'MATRIX_END'
aln.matrix = '\n'.join(aln.matrix)
continue
elif state=='ALIGN' and line[0]=='}':
state = 'ALIGN_END'
aln.chains.append(chain)
chain = None
continue
elif line[0]=='}': # End of state
continue
tokens = line.lstrip().split()
if state==None:
print line
raise Exception('Wrong')
elif state=='MATRIX':
aln.matrix.append(line)
elif state=='FILES':
block.filenames.append(tokens[0])
block.lengths.append(int(tokens[2]))
block.strands.append(strandDict[tokens[3]])
# Next line in stanza
tokens = iFile.next().strip().split()
block.filenames.append(tokens[0])
block.lengths.append(int(tokens[2]))
block.strands.append(strandDict[tokens[3]])
elif state=='HEADERS':
block.headers.append(tokens[0])
# Next line in stanza
tokens = iFile.next().strip().split()
block.headers.append(tokens[0])
elif state=='ALIGN':
if tokens[0]=='s':
chain.score = int(tokens[1])
elif tokens[0]=='b':
chain.interval1[0] = int(tokens[1])
chain.interval2[0] = int(tokens[2])
elif tokens[0]=='e':
chain.interval1[1] = int(tokens[1])
chain.interval2[1] = int(tokens[2])
elif tokens[0]=='l':
chain.hsps.append(HSP(tokens[1:]))
return aln
def __init__(self, iFileHandle1, iFileHandle2):
self.iFile1 = smartopen(iFileHandle1)
self.iFile2 = smartopen(iFileHandle2)
('-p', 'Predicted peptides only'),
('-a', email),
]
files = [('-u', '', ''), ('-v', '', '')]
try:
html = multipart.post(genomeScanURL, fields, files, proxy, proxyPort)
except multipart.FormSubmissionException:
print >> sys.stderr, "*** %s submission failed. Retry later" % description
return
except Exception, e:
print e
sys.exit('Argh!')
if oFileHandle:
oFile = smartopen(oFileHandle, 'w')
print >> oFile, html
oFile.close()
return html
def extractSeq(feature, blastDb, dx, dy):
"""Extract the translated sequence of a feature and DNA
sequence of the surrounding the genomic region.
@param feature: Feature object. Mandatory attributes: accession, sStart, sEnd.
@param blastDb: Blast database.
@param dx: Length of sequence to extract upstream.
@param dy: Length of sequence to extract downstream.
@returns: tuple of fasta strings (DNA, protein).
def __init__(self, iFileHandle):
self.iFile = smartopen(iFileHandle)
self.iFilename = self.iFile.name
self._iter = None
def load_full(iFileHandle):
"""Load genscan predictions.
Arguments:
iFileHandle -- Input file or filename.
Return values:
data -- Annotation data (a list of lists, each list in one gene)
proteins -- Predicted proteins (a list of tuples (header, sequence))
meta -- Meta-data in first 8 lines of genscan output
"""
iFile = smartopen(iFileHandle)
data = {}
proteins = []
meta = []
startPredState = '----- ---- - ------ ------ ---- -- -- ---- ---- ----- ----- ------'
endPredState = 'Predicted peptide sequence(s):'
skipState = 'Slice no. '
metaState = 'GENSCAN 1.0'
state = None
for line in iFile:
line = line.strip()
if metaState in line:
state = 'meta'
if line==startPredState:
state = 'pred'
elif line=='NO EXONS/GENES PREDICTED IN SEQUENCE':
state = 'fail'
elif line==endPredState:
state = 'prot'
elif skipState in line:
state = 'skip'
else:
if state=='meta':
if line:
meta.append(line)
elif state=='pred':
if line:
tokens = line.split()
d = Predicted(tokens)
gene = int(d.gene_exon.split('.')[0])
try:
data[gene].append(d)
except KeyError:
data[gene] = [d]
elif state=='prot':
break
elif state=='fail':
return [], [], ''
if state=='prot':
proteins = fasta.load_mfa(iFile)
data = data.items()
data.sort()
data = [x[1] for x in data]
return data, proteins, meta
