def _build_logodata(options):
motif_flag = False
fin = options.fin
if options.upload is None:
if fin is None:
fin = StringIO(sys.stdin.read())
else:
if fin is None:
from . import _from_URL_fileopen
fin = _from_URL_fileopen(options.upload)
else:
raise ValueError("error: options --fin and --upload are incompatible")
try:
# Try reading data in transfac format first.
from corebio.matrix import Motif
motif = Motif.read_transfac(fin, alphabet=options.alphabet)
motif_flag = True
except ValueError as motif_err:
# Failed reading Motif, try reading as multiple sequence data.
if options.input_parser == "transfac":
raise motif_err # Adding transfac as str insted of parser is a bit of a ugly kludge
seqs = read_seq_data(fin,
options.input_parser.read,
alphabet=options.alphabet,
ignore_lower_case=options.ignore_lower_case)
if motif_flag:
if options.ignore_lower_case:
raise ValueError("error: option --ignore-lower-case incompatible with matrix input")
if options.reverse or options.revcomp:
motif.reverse()
if options.complement or options.revcomp:
motif.complement()
prior = parse_prior(options.composition, motif.alphabet, options.weight)
data = LogoData.from_counts(motif.alphabet, motif, prior)
else:
if options.reverse or options.revcomp:
seqs = SeqList([s.reverse() for s in seqs], seqs.alphabet)
if options.complement or options.revcomp:
if not nucleic_alphabet.alphabetic(seqs.alphabet):
raise ValueError('non-nucleic sequence cannot be complemented')
aaa = seqs.alphabet
seqs.alphabet = nucleic_alphabet
seqs = SeqList([Seq(s, seqs.alphabet).complement() for s in seqs], seqs.alphabet)
seqs.alphabet = aaa
prior = parse_prior(options.composition, seqs.alphabet, options.weight)
data = LogoData.from_seqs(seqs, prior)
return data
def create_logo(self, seqs=[]):
"""Create sequence logo for input sequences."""
# seperate headers
headers, instances = [list(x)
for x in zip(*seqs)]
if self.options.sequence_type is 'rna':
alphabet = Alphabet('ACGU')
elif self.options.sequence_type is 'protein':
alphabet = Alphabet('ACDEFGHIKLMNPQRSTVWY')
else:
alphabet = Alphabet('AGCT')
motif_corebio = SeqList(alist=instances, alphabet=alphabet)
data = wbl.LogoData().from_seqs(motif_corebio)
format = wbl.LogoFormat(data, self.options)
if self.output_format == 'png':
return wbl.png_formatter(data, format)
elif self.output_format == 'png_print':
return wbl.png_print_formatter(data, format)
elif self.output_format == 'jpeg':
return wbl.jpeg_formatter(data, format)
else:
return wbl.eps_formatter(data, format)
def _build_logodata(options):
motif_flag = False
fin = options.fin
if options.upload is None:
if fin is None:
fin = StringIO(sys.stdin.read())
else:
if fin is None:
from . import _from_URL_fileopen
fin = _from_URL_fileopen(options.upload)
else:
raise ValueError(
"error: options --fin and --upload are incompatible")
try:
# Try reading data in transfac format first.
from corebio.matrix import Motif
motif = Motif.read_transfac(fin, alphabet=options.alphabet)
motif_flag = True
except ValueError as motif_err:
# Failed reading Motif, try reading as multiple sequence data.
if options.input_parser == "transfac":
raise motif_err # Adding transfac as str insted of parser is a bit of a ugly kludge
seqs = read_seq_data(fin,
options.input_parser.read,
alphabet=options.alphabet,
ignore_lower_case=options.ignore_lower_case)
if motif_flag:
if options.ignore_lower_case:
raise ValueError(
"error: option --ignore-lower-case incompatible with matrix input"
)
if options.reverse or options.revcomp:
motif.reverse()
if options.complement or options.revcomp:
motif.complement()
prior = parse_prior(options.composition, motif.alphabet,
options.weight)
data = LogoData.from_counts(motif.alphabet, motif, prior)
else:
if options.reverse or options.revcomp:
seqs = SeqList([s.reverse() for s in seqs], seqs.alphabet)
if options.complement or options.revcomp:
if not nucleic_alphabet.alphabetic(seqs.alphabet):
raise ValueError('non-nucleic sequence cannot be complemented')
aaa = seqs.alphabet
seqs.alphabet = nucleic_alphabet
seqs = SeqList([Seq(s, seqs.alphabet).complement() for s in seqs],
seqs.alphabet)
seqs.alphabet = aaa
prior = parse_prior(options.composition, seqs.alphabet, options.weight)
data = LogoData.from_seqs(seqs, prior)
return data
def read(fin, alphabet=None):
""" Extract sequence data from a nexus file."""
n = Nexus(fin)
seqs = []
for taxon in n.taxlabels:
name = safename(taxon)
r = n.matrix[taxon]
if alphabet is None:
s = Seq(r, name=name, alphabet=r.alphabet)
else:
s = Seq(r, name=name, alphabet=alphabet)
seqs.append(s)
if len(seqs) == 0:
# Something went terrible wrong.
raise ValueError("Cannot parse file")
return SeqList(seqs)
seqs = read_seq_data(fin,
options.input_parser.read,
alphabet=options.alphabet,
ignore_lower_case = options.ignore_lower_case)
if motif_flag :
if options.ignore_lower_case:
raise ValueError("error: option --ignore-lower-case incompatible with matrix input")
if options.reverse: motif.reverse()
if options.complement: motif.complement()
prior = parse_prior( options.composition,motif.alphabet, options.weight)
data = LogoData.from_counts(motif.alphabet, motif, prior)
else :
if options.reverse:
seqs = SeqList([s.reverse() for s in seqs], seqs.alphabet)
if options.complement :
seqs= SeqList( [Seq(s,seqs.alphabet).complement() for s in seqs], seqs.alphabet)
prior = parse_prior( options.composition,seqs.alphabet, options.weight)
data = LogoData.from_seqs(seqs, prior)
return data
def _build_logoformat( logodata, opts) :
motif.reverse()
if options.complement:
motif.complement()
if not isCodon:
prior, compos = parse_prior(fin_compos, motif.alphabet, fin_weight)
data = LogoData.from_counts(motif.alphabet, motif, options.stats_func, prior, compos, second_data)
else:
raise ValueError("option --sequence-type 'codon' incompatible with matrix input")
else:
if options.codon_frame < 0 and isCodon:
options.reverse = True
options.complement = True
if options.reverse:
seqs = SeqList([s.reverse() for s in seqs], seqs.alphabet)
if options.complement:
seqs = SeqList([Seq(s, seqs.alphabet).complement() for s in seqs], seqs.alphabet)
if isCodon:
if abs(options.codon_frame) > 1:
beg = abs(options.codon_frame) - 1
end = beg + int((len(seqs[0]) - beg) / 3) * 3
seqs = SeqList([s[beg:end] for s in seqs], seqs.alphabet)
if std_alphabets["dna"] == seqs.alphabet:
seqs.alphabet = codon_dna_alphabet
elif std_alphabets["rna"] == seqs.alphabet:
seqs.alphabet = codon_rna_alphabet
prior, compos = parse_prior(fin_compos, seqs.alphabet, fin_weight)
def read(fin, alphabet=None):
assert fin is not None # Do something with arguments to quite pychecker
if alphabet is not None : pass
return SeqList([])