def ab1toassembly(filename1, filename2):
"""Takes two AB1 filenames and returns an Assembly of them."""
read1 = ab1.read(filename1)
read2 = ab1.read(filename2)
return assemble(read1['sequence'], read1['confidences'], read1['traces'],
rcbases(read2['sequence']),
rcconfidences(read2['confidences']),
rctraces(read2['traces']))
def ab1toassembly(filename1, filename2):
"""Takes two AB1 filenames and returns an Assembly of them."""
read1 = ab1.read(filename1)
read2 = ab1.read(filename2)
return assemble(
read1["sequence"],
read1["confidences"],
read1["traces"],
rcbases(read2["sequence"]),
rcconfidences(read2["confidences"]),
rctraces(read2["traces"]),
)
def assemble(read1, read2, *extra_seqs):
tracks1 = ab1.read(read1)
tracks2 = ab1.read(read2)
ref = contig.contig(tracks1['sequence'], tracks1['confidences'],
tracks.revcomp(tracks2['sequence']),
tracks.revcomp(tracks2['confidences']))
t = tracks.TrackSet()
read1_offset, read1_sequence = ref['read1']
read2_offset, read2_sequence = ref['read2']
read1_confs = tracks.regap(read1_sequence, tracks1['confidences'])
read2_confs = tracks.regap(read2_sequence, tracks.revcomp(tracks2['confidences']))
read1_traces = tracks.regap(read1_sequence, tracks1['traces'])
read2_traces = tracks.regap(read2_sequence, tracks.revcomp(tracks2['traces']))
t.extend([
tracks.TrackEntry('read 1 traces', read1_offset, read1_traces),
tracks.TrackEntry('read 1 confidences', read1_offset, read1_confs),
tracks.TrackEntry('read 1 bases', read1_offset, read1_sequence),
tracks.TrackEntry('read 2 traces', read2_offset, read2_traces),
tracks.TrackEntry('read 2 confidences', read2_offset, read2_confs),
tracks.TrackEntry('read 2 bases', read2_offset, read2_sequence)])
if ref['reference'] != None:
reference_offset, reference_sequence = ref['reference']
t.append(tracks.TrackEntry('reference', reference_offset, reference_sequence))
for (name,s) in extra_seqs:
if ref['reference'] != None:
(roffset, _), (soffset, saligned) = fasta.fasta(reference_sequence, s)
t.append(tracks.TrackEntry(name, reference_offset + soffset - roffset,
tracks.sequence(saligned)))
else:
t.append(tracks.TrackEntry(name, 0, s))
# Now add an assembly of the lab sequence and reference sequence
if len(extra_seqs) == 1 and ref['reference'] != None:
labtrack = t[-1]
reftrack = t[-2]
offset = max(labtrack.offset, reftrack.offset)
loffset = offset - labtrack.offset
roffset = offset - reftrack.offset
assert loffset >= 0 and roffset >= 0 and (loffset == 0 or roffset == 0)
bases = tracks.sequence(''.join([a == b and ' ' or 'X' for a,b in
zip(labtrack.track[loffset:],
reftrack.track[roffset:])]))
if 'X' in bases:
t.append(tracks.TrackEntry('mismatches', offset, bases))
return (t, ref['strands'])
def ab1tohtml(ab1filename):
r = ab1.read(ab1filename)
a = Assembly([('traces', aflist(0, r['traces'], gap=None, trackclass='svg')),
('confidences', aflist(0, r['confidences'], gap=None,
trackclass='integer')),
('bases', aflist(0, r['sequence'], gap='-',
trackclass='nucleotide'))])
s = "<!DOCTYPE html>"
s += "<html><body>"
s += renderassembly(a)
s += "</body></html>"
return s
def ab1tohtml(ab1filename):
r = ab1.read(ab1filename)
a = Assembly([
('traces', aflist(0, r['traces'], gap=None, trackclass='svg')),
('confidences',
aflist(0, r['confidences'], gap=None, trackclass='integer')),
('bases', aflist(0, r['sequence'], gap='-', trackclass='nucleotide'))
])
s = "<!DOCTYPE html>"
s += "<html><body>"
s += renderassembly(a)
s += "</body></html>"
return s
'vent', r'\b[Bb]acteri(um|a)', r'sp\.', r'str\.'
]))
def is_unclassified(s):
"""Is string *s* containining and organism description ill posed?"""
return re.search(unclassified_regex, s) and True or False
# workup should be a dictionary with the keys "accession", "workup",
# "pat_name", "amp_name", "seq_key", as selected directly from the
# workups view of the database. In production, it will be found in a
# JSON file written in the directory.
def generate_report(lookup_fun, assembled_render, strandwise_render):
def f((workup, read1path, read2path), omit_blast=False):
read1 = ab1.read(read1path)
read2 = ab1.read(read2path)
assembly = contig.assemble(read1['sequence'], read1['confidences'],
read1['traces'], read2['sequence'],
read2['confidences'], read2['traces'])
if 'contig' in assembly:
if not omit_blast:
v = lookup_fun(''.join(assembly['contig'].values),
save_path=os.path.join(workup['path'],
'blast.xml'),
save_json=os.path.join(workup['path'],
'blast.json'))
body = assembled_render(workup, assembly, v, omit_blast=False)
else:
body = assembled_render(workup,
assembly,
r'sp\.',
r'str\.']))
def is_unclassified(s):
"""Is string *s* containining and organism description ill posed?"""
return re.search(unclassified_regex, s) and True or False
# workup should be a dictionary with the keys "accession", "workup",
# "pat_name", "amp_name", "seq_key", as selected directly from the
# workups view of the database. In production, it will be found in a
# JSON file written in the directory.
def generate_report(lookup_fun, assembled_render, strandwise_render):
def f((workup, read1path, read2path), omit_blast=False):
read1 = ab1.read(read1path)
read2 = ab1.read(read2path)
assembly = contig.assemble(read1['sequence'], read1['confidences'], read1['traces'],
read2['sequence'], read2['confidences'], read2['traces'])
if 'contig' in assembly:
if not omit_blast:
v = lookup_fun(''.join(assembly['contig'].values), save_path=os.path.join(workup['path'], 'blast.xml'),
save_json=os.path.join(workup['path'], 'blast.json'))
body = assembled_render(workup, assembly, v, omit_blast=False)
else:
body = assembled_render(workup, assembly, None, omit_blast=True)
return ('assembled', body)
else:
if not omit_blast:
v1 = lookup_fun(''.join(assembly['bases 1'].values), save_path=workup['path'],
save_json=os.path.join(workup['path'], 'blast.json'))