def records():
if refseq:
yield HMMER.valid(refseq, is_dna=is_dna)
for record in seqrecords:
if not is_dna and record.seq.alphabet == DNAAlphabet:
record = translate(record)
yield HMMER.valid(record, is_dna=is_dna)
def records():
if refseq:
yield HMMER.valid(refseq, is_dna=is_dna)
for record in seqrecords:
if not is_dna and record.seq.alphabet == DNAAlphabet:
record = translate(record)
yield HMMER.valid(record, is_dna=is_dna)
def generate_hmm_(opts):
fd, tmphmm = mkstemp()
close(fd)
fd, tmpaln = mkstemp()
close(fd)
is_dna = opts.ENCODER == DNAEncoder
try:
with open(opts.REFMSA) as msa_fh:
with open(tmpaln, 'w') as aln_fh:
msa_fmt = seqfile_format(opts.REFMSA)
source = Verifier(SeqIO.parse(msa_fh, msa_fmt), DNAAlphabet)
try:
SeqIO.write((record if is_dna else translate(record)
for record in source), aln_fh, 'stockholm')
except VerifyError:
if is_dna:
raise RuntimeError(
"DNA encoding incompatible with protein reference MSA"
)
source.set_alphabet(AminoAlphabet)
aln_fh.seek(0)
SeqIO.write(source, aln_fh, 'stockholm')
hmmer = HMMER(opts.HMMER_ALIGN_BIN, opts.HMMER_BUILD_BIN)
hmmer.build(tmphmm,
tmpaln,
alphabet=HMMER.DNA if is_dna else HMMER.AMINO)
finally:
if exists(tmpaln):
remove(tmpaln)
return tmphmm
def generate_alignment_(seqrecords, hmmfile, opts, refseq=None):
fd, tmpseq = mkstemp(); close(fd)
fd, tmpaln = mkstemp(); close(fd)
finished = False
is_dna = opts.ENCODER == DNAEncoder
log = getLogger(IDEPI_LOGGER)
try:
# get the FASTA format file so we can HMMER it
with open(tmpseq, 'w') as seq_fh:
def records():
if refseq:
yield HMMER.valid(refseq, is_dna=is_dna)
for record in seqrecords:
if not is_dna and record.seq.alphabet == DNAAlphabet:
record = translate(record)
yield HMMER.valid(record, is_dna=is_dna)
SeqIO.write(records(), seq_fh, 'fasta')
log.debug('aligning sequences')
hmmer = HMMER(opts.HMMER_ALIGN_BIN, opts.HMMER_BUILD_BIN)
hmmer.align(
hmmfile,
tmpseq,
output=tmpaln,
alphabet=HMMER.DNA if is_dna else HMMER.AMINO,
outformat=HMMER.PFAM
)
# rename the final alignment to its destination
finished = True
finally:
# cleanup these files
if exists(tmpseq):
remove(tmpseq)
if not finished:
raise RuntimeError("failed to generate alignment")
return tmpaln
def generate_alignment_(seqrecords, hmmfile, opts, refseq=None):
fd, tmpseq = mkstemp()
close(fd)
fd, tmpaln = mkstemp()
close(fd)
finished = False
is_dna = opts.ENCODER == DNAEncoder
log = getLogger(IDEPI_LOGGER)
try:
# get the FASTA format file so we can HMMER it
with open(tmpseq, 'w') as seq_fh:
def records():
if refseq:
yield HMMER.valid(refseq, is_dna=is_dna)
for record in seqrecords:
if not is_dna and record.seq.alphabet == DNAAlphabet:
record = translate(record)
yield HMMER.valid(record, is_dna=is_dna)
SeqIO.write(records(), seq_fh, 'fasta')
log.debug('aligning sequences')
hmmer = HMMER(opts.HMMER_ALIGN_BIN, opts.HMMER_BUILD_BIN)
hmmer.align(hmmfile,
tmpseq,
output=tmpaln,
alphabet=HMMER.DNA if is_dna else HMMER.AMINO,
outformat=HMMER.PFAM)
# rename the final alignment to its destination
finished = True
finally:
# cleanup these files
if exists(tmpseq):
remove(tmpseq)
if not finished:
raise RuntimeError("failed to generate alignment")
return tmpaln
def generate_hmm_(opts):
fd, tmphmm = mkstemp(); close(fd)
fd, tmpaln = mkstemp(); close(fd)
is_dna = opts.ENCODER == DNAEncoder
try:
with open(opts.REFMSA) as msa_fh:
with open(tmpaln, 'w') as aln_fh:
msa_fmt = seqfile_format(opts.REFMSA)
source = Verifier(SeqIO.parse(msa_fh, msa_fmt), DNAAlphabet)
try:
SeqIO.write(
(record if is_dna else translate(record) for record in source),
aln_fh,
'stockholm')
except VerifyError:
if is_dna:
raise RuntimeError("DNA encoding incompatible with protein reference MSA")
source.set_alphabet(AminoAlphabet)
aln_fh.seek(0)
SeqIO.write(
source,
aln_fh,
'stockholm')
hmmer = HMMER(opts.HMMER_ALIGN_BIN, opts.HMMER_BUILD_BIN)
hmmer.build(
tmphmm,
tmpaln,
alphabet=HMMER.DNA if is_dna else HMMER.AMINO
)
finally:
if exists(tmpaln):
remove(tmpaln)
return tmphmm