def calc_sample_coverage(args):
"""Counts the total number of non-gap/ambiguous characters for
each sample per contig.
"""
mvf = MultiVariantFile(args.mvf, 'read')
data = {}
# Set up sample indices
sample_labels = mvf.get_sample_labels()
if args.sample_indices is not None:
sample_indices = [int(x) for x in
args.sample_indices[0].split(",")]
elif args.sample_labels is not None:
sample_indices = mvf.get_sample_indices(
labels=args.sample_labels[0].split(","))
else:
sample_indices = mvf.get_sample_indices()
# Set up contig ids
if args.contig_ids is not None:
contig_ids = args.contig_ids[0].split(",")
elif args.contig_labels is not None:
contig_ids = mvf.get_contig_ids(
labels=args.contig_labels[0].split(","))
else:
contig_ids = mvf.get_contig_ids()
for contig, _, allelesets in mvf.iterentries(
contigs=contig_ids, subset=sample_indices,
decode=True):
if contig not in data:
data[contig] = dict.fromkeys(sample_labels, 0)
data[contig]['contig'] = contig
for j, x in enumerate(sample_indices):
data[contig][sample_labels[x]] += int(
allelesets[0][j] not in 'Xx-')
outfile = OutputFile(path=args.out,
headers=(["contig"] + [sample_labels[x] for x in
sample_indices]))
for contig in data:
outfile.write_entry(data[contig])
return ''
def mvf2phy(args):
"""Main method"""
mvf = MultiVariantFile(args.mvf, 'read')
if (mvf.flavor in ("dna", "rna") and args.output_data == "prot") or (
mvf.flavor == "prot" and args.output_data in ("dna", "rna")):
raise RuntimeError(
"--outdput-data {} incompatiable with '{}' flavor mvf".format(
args.output_data, mvf.flavor))
max_region_coord = dict((x, None) for x in mvf.get_contig_ids())
if args.regions is not None:
_, max_region_coord, _ = parse_regions_arg(args.regions,
mvf.get_contig_ids())
if args.sample_indices is not None:
sample_indices = [int(x) for x in args.sample_indices[0].split(",")]
elif args.sample_labels is not None:
sample_indices = mvf.get_sample_indices(
ids=args.sample_labels[0].split(","))
else:
sample_indices = mvf.get_sample_indices()
sample_labels = mvf.get_sample_ids(indices=sample_indices)
skipcontig = ''
tmp_files = dict((fn,
open("{}-{}.tmp".format(fn, randint(1000000, 9999999)),
'w+', args.buffer)) for fn in sample_labels)
labelwritten = dict.fromkeys(sample_labels, False)
current_contig_id = None
current_contig_start = 1
current_contig_end = 1
if args.partition is True:
partprefix = "PROT" if args.output_data == "prot" else "DNA"
partitionfile = open("{}.part".format(args.out), 'w')
for contig, _, allelesets in mvf.iterentries(
contig_ids=(mvf.get_contig_ids()
if args.regions is None else max_region_coord[:]),
decode=True):
if contig == skipcontig:
continue
if contig not in max_region_coord:
skipcontig = contig[:]
continue
if current_contig_id is None:
current_contig_id = contig[:]
elif contig != current_contig_id:
if args.partition is True:
if current_contig_end > current_contig_start:
partitionfile.write("{}, {} = {}-{}\n".format(
partprefix,
mvf.get_contig_labels(ids=current_contig_id),
current_contig_start, current_contig_end - 1))
current_contig_id = contig[:]
# reset start as one position after end of last
current_contig_start = current_contig_end
current_contig_end = current_contig_end + 1
for col, label in zip(sample_indices, sample_labels):
if not labelwritten[label]:
if args.label_type == 'long':
tmp_files[label].write("{}{}".format(
label[:100], " " * (100 - len(label[:100]))))
elif args.label_type == 'short':
tmp_files[label].write("{}{}".format(
label[:20], " " * (20 - len(label[:20]))))
labelwritten[label] = True
if mvf.flavor == 'dna':
tmp_files[label].write(allelesets[0][col] == 'X' and 'N'
or allelesets[0][col])
if label == sample_labels[0]:
current_contig_end += 1
elif ((mvf.flavor == 'codon' and args.output_data == 'prot')
or (mvf.flavor == 'prot')):
tmp_files[label].write(allelesets[0][col])
if label == sample_labels[0]:
current_contig_end += 1
elif mvf.flavor == 'codon':
codon = [
"N" if allelesets[x][col] == 'X' else allelesets[x][col]
for x in (1, 2, 3)
]
tmp_files[label].write(''.join(codon))
if label == sample_labels[0]:
current_contig_end += 3
first_file = True
totalseqlen = 0
with open(args.out, 'w') as outfile:
for filehandler in tmp_files.values():
# read first file to establish sequence length for phylip header
if first_file is True:
filehandler.seek(0, 0)
buff = filehandler.read(args.buffer)
while buff != '':
if " " in buff:
totalseqlen += len(buff.strip().split(" ")[-1])
else:
totalseqlen += len(buff.strip())
buff = filehandler.read(args.buffer)
outfile.write("{} {}\n".format(len(sample_labels),
totalseqlen))
first_file = False
filehandler.seek(0, 0)
buff = filehandler.read(args.buffer)
while buff != '':
if first_file is True:
outfile.write("{} {}\n".format(len(sample_labels),
len(buff.split()[1])))
first_file = False
outfile.write(buff)
buff = filehandler.read(args.buffer)
outfile.write("\n")
filehandler.close()
os.remove(os.path.join(args.temp_dir, filehandler.name))
if args.partition is True:
if current_contig_end > current_contig_start:
partitionfile.write("{}, {} = {}-{}\n".format(
partprefix, mvf.get_contig_labels(ids=current_contig_id),
current_contig_start, current_contig_end - 1))
partitionfile.close()
return ''
def vcf2mvf(args=None):
"""Main method for vcf2mvf"""
sepchars = dict([("TAB", "\t"), ("SPACE", " "), ("DBLSPACE", " "),
("COMMA", ","), ("MIXED", None)])
args.fieldsep = sepchars[args.field_sep]
# ESTABLISH VCF
args.qprint("Opening input VCF: {}".format(args.vcf))
vcf = VariantCallFile(args.vcf, indexcontigs=(not args.no_autoindex))
# ESTABLISH MVF
args.qprint("Establishing output MVF: {}".format(args.out))
mvf = MultiVariantFile(args.out, 'write', overwrite=args.overwrite)
mvf.notes.append(args.command_string)
mvf.metadata['mvfversion'] = args.versionx
# PROCESS CONTIG INFO
args.qprint("Processing VCF headers.")
vcfcontigs = vcf.metadata['contigs'].copy()
args.qprint("{} contigs found.".format(len(vcfcontigs)))
contig_translate = {}
if args.contig_ids:
for cid, cvcf, cmvf in (x.split(';') for x in args.contig_ids):
try:
cid = int(cid)
except ValueError:
pass
assert cvcf in [vcfcontigs[x]['label'] for x in vcfcontigs]
for vid in vcfcontigs:
if vcfcontigs[vid]['label'] == cvcf:
contig_translate[cvcf] = [cid, cmvf]
if cid in mvf.metadata['contigs']:
raise RuntimeError(
'Contig id {} is not unique'.format(cid))
mvf.metadata['contigs'][cid] = vcfcontigs[vid].copy()
if cmvf in mvf.get_contig_labels():
raise RuntimeError(
'Contig label {} is not unique'.format(cmvf))
mvf.metadata['contigs'][cid]['label'] = cmvf[:]
mvf.reset_max_contig()
mvf.max_contig_index -= 1
args.qprint("Processing contigs.")
static_contig_ids = list(mvf.get_contig_ids())
for vcid in vcfcontigs:
vlabel = vcfcontigs[vcid]['label']
if vlabel not in static_contig_ids:
newindex = mvf.get_next_contig_index()
if ((is_int(vlabel) or len(vlabel) < 3)
and vlabel not in static_contig_ids):
newid = vlabel[:]
else:
newid = str(newindex)
mvf.contig_indices.append(newindex)
mvf.contig_ids.append(newid)
mvf.contig_data[newindex] = vcfcontigs[vcid].copy()
static_contig_ids.append(newid)
contig_translate[vlabel] = [newindex, vlabel]
mvf.reset_max_contig()
new_contigs = [(x, mvf.contig_data[x]['label'])
for x in mvf.contig_indices]
if args.skip_contig_label_check is False:
args.qprint("Checking contigs for label/id overlap errors.")
xids = [x[0] for x in new_contigs]
xlabels = [x[1] for x in new_contigs]
xintersect = set(xids).intersection(xlabels)
if xintersect:
for i, (newid, newlabel) in enumerate(new_contigs):
if i % 100 == 0:
args.qprint("{} contigs processed".format(i))
if newid in xlabels[:i] or newid in xlabels[i + 1:]:
# if newid in xlabels:
# if xlabels.index(newid) != i:
raise RuntimeError("Error contig id {} is the same as"
" the label for another contig"
" ({})".format(newid,
xlabels.index(newid)))
if newlabel in xids[:i] or newlabel in xids[i + 1:]:
# if newlabel in xids:
# if xids.index(newlabel) != i:
raise RuntimeError("Error contig label {} is the same"
"as the id for another contig"
"({})".format(newlabel,
xids.index(newlabel)))
# PROCESS SAMPLE INFO
args.qprint("Processing samples.")
samplelabels = [args.ref_label] + vcf.metadata['samples'][:]
if args.alleles_from:
args.alleles_from = args.alleles_from.split(':')
samplelabels += args.alleles_from
if args.sample_replace:
newsample = [
x.split(':') if ':' in tuple(x) else tuple([x, x])
for x in args.sample_replace
]
unmatched = list(enumerate(samplelabels))
for old, new in newsample:
labelmatched = False
for j, (i, name) in enumerate(unmatched):
if old in name:
samplelabels[i] = new
labelmatched = j
break
if labelmatched is not False:
del unmatched[labelmatched]
mvf.sample_indices = list(range(len(samplelabels)))
mvf.sample_ids = samplelabels[:]
for i, label in enumerate(samplelabels):
mvf.sample_data[i] = {'id': label}
mvf.metadata['ncol'] = len(mvf.sample_ids)
mvf.max_sample_index = len(mvf.sample_ids)
mvf.metadata['sourceformat'] = vcf.metadata['sourceformat']
# WRITE MVF HEADER
mvf.write_data(mvf.get_header())
mvfentries = []
nentry = 0
args.qprint("Processing VCF entries.")
for vcfrecord in vcf.iterentries(args):
mvfstring = ''.join(vcfrecord['genotypes'])
if args.filter_nonref_empty is True:
if all(x in 'Xx-?' for x in mvfstring[1:]):
continue
mvf_alleles = encode_mvfstring(mvfstring)
if args.out_flavor in ('dnaqual', ):
qual_alleles = encode_mvfstring(''.join(vcfrecord['qscores']))
if mvf_alleles:
mvfentries.append(
(contig_translate.get(vcfrecord['contig'])[0],
vcfrecord['coord'],
((mvf_alleles,
qual_alleles) if args.out_flavor in ('dnaqual', ) else
(mvf_alleles, ))))
nentry += 1
if nentry == args.line_buffer:
mvf.write_entries(mvfentries, encoded=True)
mvfentries = []
nentry = 0
if mvfentries:
mvf.write_entries(mvfentries)
mvfentries = []
return ''
def calc_all_character_count_per_sample(args):
"""Count the number of and relative rate of certain bases
spatially along chromosomes
"""
args.qprint("Running CalcAllCharacterCountPerSample")
mvf = MultiVariantFile(args.mvf, 'read')
current_contig = None
current_position = 0
data_in_buffer = False
# Set up sample indices
sample_labels = mvf.get_sample_ids()
if args.sample_indices is not None:
sample_indices = [int(x) for x in args.sample_indices[0].split(",")]
elif args.sample_labels is not None:
sample_indices = mvf.get_sample_indices(
ids=args.sample_labels[0].split(","))
else:
sample_indices = mvf.get_sample_indices()
# Set up contig ids
if args.contig_ids is not None:
contig_ids = args.contig_ids[0].split(",")
elif args.contig_labels is not None:
contig_ids = mvf.get_contig_ids(
labels=args.contig_labels[0].split(","))
else:
contig_ids = None
data = dict((i, {}) for i in sample_indices)
data_characters = [{} for i in sample_indices]
for contig, pos, allelesets in mvf.iterentries(decode=False,
contig_ids=contig_ids):
# Check Minimum Site Coverage
if check_mincoverage(args.mincoverage, allelesets[0]) is False:
continue
if current_contig is None:
current_contig = contig[:]
if args.windowsize > 0:
while pos > current_position + args.windowsize - 1:
current_position += args.windowsize
# Check if windows are specified.
if not same_window((current_contig, current_position),
(contig, pos), args.windowsize):
args.qprint("Processing contig {}".format(current_contig))
for i in sample_indices:
data[i][(current_contig, current_position)] = {
'contig': current_contig,
'position': current_position
}
data[i][(current_contig,
current_position)].update(data_characters[i])
if contig != current_contig:
current_contig = contig[:]
current_position = 0
else:
current_position += (0 if args.windowsize == -1 else
args.windowsize)
data_characters = [{} for i in sample_indices]
data_in_buffer = False
alleles = allelesets[0]
if len(alleles) == 1:
for i in sample_indices:
data_characters[i][alleles[0]] = (
data_characters[i].get(alleles[0], 0) + 1)
else:
alleles = mvf.decode(alleles)
for i in sample_indices:
data_characters[i][alleles[i]] = (
data_characters[i].get(alleles[i], 0) + 1)
data_in_buffer = True
if data_in_buffer:
for i in sample_indices:
data[i][(current_contig, current_position)] = {
'contig': current_contig,
'position': current_position
}
data[i][(current_contig,
current_position)].update(data_characters[i])
# WRITE OUTPUT
all_chars = set([])
for sampleid in data:
for window in data[sampleid]:
all_chars.update([
x for x in data[sampleid][window]
if x not in ('contig', 'position')
])
headers = ['contig', 'position']
headers.extend(list(sorted(all_chars)))
outfile = OutputFile(path=args.out, headers=headers)
for sampleid in sample_indices:
outfile.write("#{}\n".format(sample_labels[sampleid]))
sorted_entries = [(data[sampleid][k]['contig'],
data[sampleid][k]['position'], k)
for k in data[sampleid]]
for _, _, k in sorted_entries:
outfile.write_entry(data[sampleid][k], defaultvalue='0')
return ''
def calc_dstat_combinations(args):
"""Calculate genome-wide D-statstics for
all possible trio combinations of samples
and outgroups specified.
"""
mvf = MultiVariantFile(args.mvf, 'read')
data = {}
sample_labels = mvf.get_sample_ids()
if args.outgroup_indices is not None:
outgroup_indices = [
int(x) for x in args.outgroup_indices[0].split(",")
]
elif args.outgroup_labels is not None:
outgroup_indices = mvf.get_sample_indices(
ids=args.outgroup_labels[0].split(","))
if args.sample_indices is not None:
sample_indices = [int(x) for x in args.sample_indices[0].split(",")]
elif args.sample_labels is not None:
sample_indices = mvf.get_sample_indices(
ids=args.sample_labels[0].split(","))
else:
sample_indices = mvf.get_sample_indices()
if args.contig_ids is not None:
contig_ids = args.contig_ids[0].split(",")
elif args.contig_labels is not None:
contig_ids = mvf.get_contig_ids(
labels=args.contig_labels[0].split(","))
else:
contig_ids = None
if any(x in outgroup_indices for x in sample_indices):
raise RuntimeError("Sample and Outgroup column lists cannot overlap.")
for contig, _, allelesets in mvf:
if contig not in contig_ids:
continue
alleles = mvf.decode(allelesets[0])
for i, j, k in combinations(sample_indices, 3):
for outgroup in outgroup_indices:
subset = [alleles[x] for x in [i, j, k, outgroup]]
if any(x not in 'ATGC' for x in subset):
continue
if subset[-1] not in subset[:3]:
continue
if len(set(subset)) != 2:
continue
# [ABBA, BABA, BBAA]
val = (0 + 1 * (subset[0] == subset[3]) + 2 *
(subset[1] == subset[3]) + 4 * (subset[2] == subset[3]))
if val in (1, 2):
val -= 1
elif val == 4:
val = 2
else:
continue
tetrad = (i, j, k, outgroup)
if tetrad not in data:
data[tetrad] = {}
if contig not in data[tetrad]:
data[tetrad][contig] = [0, 0, 0]
data[tetrad][contig][val] += 1
# WRITE OUTPUT
headers = ['sample0', 'sample1', 'sample2', "outgroup"]
for xcontig in contig_ids:
headers.extend([
'{}:abba'.format(xcontig), '{}:baba'.format(xcontig),
'{}:bbaa'.format(xcontig), '{}:D'.format(xcontig)
])
outfile = OutputFile(path=args.out, headers=headers)
for i, j, k in combinations(sample_indices, 3):
for outgroup in outgroup_indices:
tetrad = tuple([i, j, k, outgroup])
if tetrad not in data:
continue
entry = dict(('sample{}'.format(i), sample_labels[x])
for i, x in enumerate(tetrad[:3]))
entry['outgroup'] = sample_labels[outgroup]
for contig in contig_ids:
if contig not in data[tetrad]:
entry.update(dict().fromkeys([
'{}:abba'.format(contig), '{}:baba'.format(contig),
'{}:bbaa'.format(contig), '{}:D'.format(contig)
], '0'))
else:
[abba, baba, bbaa] = data[tetrad][contig]
if abba > baba and abba > bbaa:
dstat = zerodiv(baba - bbaa, baba + bbaa)
elif baba > bbaa and baba > abba:
dstat = zerodiv(abba - bbaa, abba + bbaa)
else:
dstat = zerodiv(abba - baba, abba + baba)
entry.update([('{}:abba'.format(contig), abba),
('{}:baba'.format(contig), baba),
('{}:bbaa'.format(contig), bbaa),
('{}:D'.format(contig), dstat)])
outfile.write_entry(entry)
return ''
def plot_chromoplot(args):
"""Main method"""
pallette = Pallette()
if args.colors is not None:
pallette.basecolors = args.colors
# Establish MVF and parse chromosome information
if args.quiet is False:
print("Reading MVF...")
mvf = MultiVariantFile(args.mvf, 'read')
if args.quiet is False:
print("Parsing headers...")
if args.contig_ids is not None:
contigids = args.contig_ids[0].split(",")
elif args.contig_labels is not None:
contigids = mvf.get_contig_ids(labels=args.contig_labels[0].split(","))
else:
contigids = mvf.get_contig_ids()
if args.quiet is False:
print("Plotting chromoplot for contigs: {}".format(
",".join(contigids)))
sample_labels = mvf.get_sample_labels()
if args.sample_indices is not None:
sample_indices = [int(x) for x in args.sample_indices[0].split(",")]
elif args.sample_labels is not None:
sample_indices = mvf.get_sample_indices(
labels=args.sample_labels[0].split(","))
else:
sample_indices = mvf.get_sample_indices()
assert len(sample_indices) >= 3
if args.outgroup_indices is not None:
outgroup_indices = [
int(x) for x in args.outgroup_indices[0].split(",")
]
elif args.outgroup_labels is not None:
outgroup_indices = mvf.get_sample_indices(
labels=args.outgroup_labels[0].split(","))
assert len(outgroup_indices) >= 1
quartets = [(x, y, z, outgroup)
for x, y, z in combinations(sample_indices, 3)
for outgroup in outgroup_indices]
# Begin iterations
for quartet_indices in quartets:
quartet_labels = [sample_labels[x] for x in quartet_indices]
if args.quiet is False:
print("Beginning quartet {}".format(",".join(quartet_labels)))
params = {
'contigs': [[
contigid, mvf.metadata['contigs'][contigid]['label'],
mvf.metadata['contigs'][contigid]['length']
] for contigid in contigids],
'outpath':
((args.out_prefix if args.out_prefix is not None else '')
or '_'.join(quartet_labels)) + ".png",
'labels':
quartet_labels,
'indices':
quartet_indices,
'windowsize':
args.windowsize,
'majority':
args.majority,
'infotrack':
args.info_track,
'yscale':
args.yscale,
'xscale':
args.xscale,
'quiet':
args.quiet,
'plottype':
args.plot_type
}
chromoplot = Chromoplot(params=params, pallette=pallette)
current_contig = ''
for contig, pos, allelesets in mvf.iterentries(subset=quartet_indices,
decode=True,
contigs=contigids):
if contig != current_contig:
if args.quiet is False:
print("Starting contig {}".format(contig))
current_contig = contig[:]
alleles = allelesets[0]
if '-' in alleles:
site_code = 'gap'
elif any(x not in 'ATGCatgc' for x in alleles):
site_code = 'ambiguous'
elif alleles[3] not in alleles[:3]:
site_code = 'nonpolar'
elif len(set(alleles)) > 2:
site_code = 'triallelic'
else:
site_code = sum([
2**(3 - j) * (alleles[j] != alleles[3]) for j in range(3)
])
chromoplot.add_data(str(contig), int(pos // args.windowsize),
site_code)
contig = ''
current_contig = ''
if not args.quiet:
print("Writing image...")
chromoplot.plot_chromoplot()
if not args.quiet:
print("Writing log...")
chromoplot.write_total_log()
return ''
def calc_character_count(args):
"""Count the number of and relative rate of certain bases
spatially along chromosomes
"""
mvf = MultiVariantFile(args.mvf, 'read')
data = {}
current_contig = None
current_position = 0
all_match = 0
all_total = 0
data_in_buffer = False
# Set up base matching from special words
def proc_special_word(argx):
if argx == 'dna':
argx = MLIB.validchars['dna']
elif argx == 'dnaambig2':
argx = MLIB.validchars['dna+ambig2']
elif argx == 'dnaambig3':
argx = MLIB.validchars['dna+ambig3']
elif argx == 'dnaambigall':
argx = MLIB.validchars['dna+ambigall']
elif argx == 'prot':
argx = MLIB.validchars['amino']
return argx
args.base_match = proc_special_word(args.base_match)
args.base_total = proc_special_word(args.base_total)
# Set up sample indices
sample_labels = mvf.get_sample_labels()
if args.sample_indices is not None:
sample_indices = [int(x) for x in args.sample_indices[0].split(",")]
elif args.sample_labels is not None:
sample_indices = mvf.get_sample_indices(
labels=args.sample_labels[0].split(","))
else:
sample_indices = mvf.get_sample_indices()
# Set up contig ids
if args.contig_ids is not None:
contig_ids = args.contig_ids[0].split(",")
elif args.contig_labels is not None:
contig_ids = mvf.get_contig_ids(
labels=args.contig_labels[0].split(","))
else:
contig_ids = None
match_counts = dict().fromkeys([sample_labels[i] for i in sample_indices],
0)
total_counts = dict().fromkeys([sample_labels[i] for i in sample_indices],
0)
for contig, pos, allelesets in mvf.iterentries(decode=False,
contigs=contig_ids):
# Check Minimum Site Coverage
if check_mincoverage(args.mincoverage, allelesets[0]) is False:
continue
#if contig not in contig_ids:
# continue
# Establish first contig
if current_contig is None:
current_contig = contig[:]
if args.windowsize > 0:
while pos > current_position + args.windowsize - 1:
current_position += args.windowsize
# Check if windows are specified.
if not same_window((current_contig, current_position),
(contig, pos), args.windowsize):
data[(current_contig, current_position)] = {
'contig': current_contig,
'position': current_position
}
for k in match_counts:
data[(current_contig, current_position)].update([
(k + '.match', match_counts[k] + all_match),
(k + '.total', total_counts[k] + all_total),
(k + '.prop', ((float(match_counts[k] + all_match) /
float(total_counts[k] + all_total))
if total_counts[k] + all_total > 0 else 0))
])
if contig != current_contig:
current_contig = contig[:]
current_position = 0
else:
current_position += (0 if args.windowsize == -1 else
args.windowsize)
match_counts = dict().fromkeys(
[sample_labels[i] for i in sample_indices], 0)
total_counts = dict().fromkeys(
[sample_labels[i] for i in sample_indices], 0)
all_total = 0
all_match = 0
data_in_buffer = False
else:
alleles = allelesets[0]
if len(alleles) == 1:
if args.base_match is None:
all_match += 1
elif alleles in args.base_match:
all_match += 1
if args.base_total is None:
all_total += 1
elif alleles in args.base_total:
all_total += 1
else:
alleles = mvf.decode(alleles)
for i in sample_indices:
if args.base_match is None:
match_counts[sample_labels[i]] += 1
elif alleles[i] in args.base_match:
match_counts[sample_labels[i]] += 1
if args.base_total is None:
total_counts[sample_labels[i]] += 1
elif alleles[i] in args.base_total:
total_counts[sample_labels[i]] += 1
data_in_buffer = True
if data_in_buffer:
data[(current_contig, current_position)] = {
'contig': current_contig,
'position': current_position
}
for k in match_counts:
data[(current_contig, current_position)].update([
(k + '.match', match_counts[k] + all_match),
(k + '.total', total_counts[k] + all_total),
(k + '.prop', ((float(match_counts[k] + all_match) /
float(total_counts[k] + all_total))
if total_counts[k] + all_total > 0 else 0))
])
# WRITE OUTPUT
headers = ['contig', 'position']
for label in sample_labels:
headers.extend([label + x for x in ('.match', '.total', '.prop')])
outfile = OutputFile(path=args.out, headers=headers)
sorted_entries = sorted([(data[k]['contig'], data[k]['position'], k)
for k in data])
for _, _, k in sorted_entries:
outfile.write_entry(data[k])
return ''
def translate_mvf(args):
"""Main method"""
args.qprint("Running TranslateMVF")
if args.gff:
args.qprint("Reading and Indexing MVF.")
else:
args.qprint("Reading MVF.")
mvf = MultiVariantFile(args.mvf, 'read', contigindex=bool(args.gff))
if mvf.flavor != 'dna':
raise RuntimeError("MVF must be flavor=dna to translate")
if args.gff:
args.qprint("Processing MVF Index File.")
mvf.read_index_file()
args.qprint("GFF processing start.")
gff_genes, gene_order = parse_gff_exome(args)
args.qprint("GFF processed.")
outmvf = MultiVariantFile(args.out, 'write', overwrite=args.overwrite)
outmvf.copy_headers_from(mvf)
outmvf.contig_data = dict(
(
i, dict((y, z)
for (y, z) in gff_genes[x].items()
if y not in ('cds', )))
for (i, x) in enumerate(gene_order))
outmvf.contig_indices = list(range(len(gene_order)))
outmvf.contig_ids = [gff_genes[x]['id']
for x in gene_order]
outmvf.contig_labels = [gff_genes[x]['label']
for x in gene_order]
outmvf.flavor = args.output_data
outmvf.metadata.notes.append(args.command_string)
outmvf.write_data(outmvf.get_header())
args.qprint("Output MVF Established.")
entrybuffer = []
nentry = 0
pos = None
if not args.gff:
args.qprint("No GFF used, translating sequences as pre-aligned in "
"coding frame.")
inputbuffer = []
current_contig = ''
for contigid, pos, allelesets in mvf.iterentries(decode=False):
if current_contig == '':
current_contig = contigid[:]
if contigid == current_contig:
inputbuffer.append((pos, allelesets))
else:
for _, amino_acids, alleles in iter_codons(
inputbuffer, mvf):
if all([x in '-X' for x in amino_acids]):
continue
if args.output_data == 'protein':
entrybuffer.append(
(current_contig, pos, (amino_acids,)))
else:
entrybuffer.append((
current_contig, pos, (
amino_acids, alleles[0],
alleles[1], alleles[2])))
nentry += 1
if nentry == args.line_buffer:
outmvf.write_entries(entrybuffer)
entrybuffer = []
nentry = 0
inputbuffer = [(pos, allelesets)]
current_contig = contigid[:]
if inputbuffer:
for _, amino_acids, alleles in iter_codons(
inputbuffer, outmvf):
if all([x in '-X' for x in amino_acids]):
continue
if args.output_data == 'protein':
entrybuffer.append(
(current_contig, pos, (amino_acids,)))
else:
entrybuffer.append((
current_contig, pos, (
amino_acids, alleles[0],
alleles[1], alleles[2])))
nentry += 1
if nentry == args.line_buffer:
outmvf.write_entries(entrybuffer)
entrybuffer = []
nentry = 0
else:
running_gene_index = -1
for igene, gene in enumerate(gene_order):
xcontiglabel = gff_genes[gene]['contig']
xcontig = mvf.get_contig_indices(
labels=gff_genes[gene]['contig'])
if xcontig is None:
print("Warning: contig {} not found".format(
gff_genes[gene]['contig']))
xcontigid = mvf.get_contig_ids(indices=xcontig)[0]
min_gene_coord = gff_genes[gene]['cds'][0][0]
max_gene_coord = gff_genes[gene]['cds'][-1][1]
mvf_entries = {}
if not igene % 100:
args.qprint("Processing gene {} on {}".format(
gene, xcontiglabel))
for contigid, pos, allelesets in mvf.itercontigentries(
xcontig, decode=False):
if pos < min_gene_coord:
continue
if pos > max_gene_coord:
break
mvf_entries[pos] = allelesets[0]
reverse_strand = gff_genes[gene]['strand'] == '-'
coords = []
running_gene_index += 1
for elem in gff_genes[gene]['cds']:
coords.extend(list(range(elem[0], elem[1] + 1)))
if reverse_strand:
coords = coords[::-1]
for codoncoord in range(0, len(coords), 3):
alleles = tuple(mvf_entries.get(x, '-')
for x in coords[codoncoord:codoncoord + 3])
if len(alleles) < 3:
alleles = tuple(list(alleles) + ['-'] * (3 - len(alleles)))
if all(len(x) == 1 for x in alleles):
if reverse_strand:
alleles = tuple(
MLIB.complement_bases[x] for x in alleles)
decoded_alleles = alleles
amino_acids = translate_single_codon(''.join(alleles))
else:
if reverse_strand is True:
decoded_alleles = tuple(tuple(MLIB.complement_bases[y]
for y in mvf.decode(x))
for x in alleles)
alleles = tuple(outmvf.encode(''.join(x))
for x in decoded_alleles)
else:
decoded_alleles = tuple(mvf.decode(x) for x in alleles)
amino_acids = tuple(translate_single_codon(''.join(x))
for x in zip(*decoded_alleles))
amino_acids = outmvf.encode(''.join(amino_acids))
if args.output_data == 'protein':
entrybuffer.append((
(
xcontigid
if args.retain_contigs
else running_gene_index
),
(
coords[codoncoord]
if args.retain_coords
else codoncoord
),
(
amino_acids,
)
))
elif args.output_data == 'codon':
entrybuffer.append((
(
xcontigid
if args.retain_contigs
else running_gene_index
),
(
coords[codoncoord]
if args.retain_coords
else codoncoord
),
(
amino_acids,
alleles[0],
alleles[1],
alleles[2]
)
))
elif args.output_data == 'dna':
for j, elem in enumerate(
range(codoncoord,
min(codoncoord + 3, len(coords)))):
entrybuffer.append((
(
xcontigid
if args.retain_contigs
else running_gene_index
),
(
coords[elem]
if args.retain_coords
else elem + 1
),
(
alleles[j],
)
))
nentry += 1
if nentry >= args.line_buffer:
args.qprint("Writing a block of {} entries.".format(
args.line_buffer))
outmvf.write_entries(entrybuffer)
entrybuffer = []
nentry = 0
if entrybuffer:
outmvf.write_entries(entrybuffer)
entrybuffer = []
nentry = 0
return ''
def legacy_translate_mvf(args):
"""Main method"""
args.qprint("Running LegacyTranslateMVF")
if args.gff:
args.qprint("Reading and Indexing MVF.")
else:
args.qprint("Reading MVF.")
mvf = MultiVariantFile(args.mvf, 'read', contigindex=bool(args.gff))
if mvf.flavor != 'dna':
raise RuntimeError("MVF must be flavor=dna to translate")
if args.gff:
args.qprint("Processing MVF Index File.")
mvf.read_index_file()
args.qprint("GFF processing start.")
gff = parse_gff_legacy_translate(
args.gff, args,
parent_gene_pattern=args.parent_gene_pattern)
args.qprint("GFF processed.")
outmvf = MultiVariantFile(args.out, 'write', overwrite=args.overwrite)
outmvf.copy_headers_from(mvf)
outmvf.flavor = args.output_data
outmvf.write_data(outmvf.get_header())
args.qprint("Output MVF Established.")
entrybuffer = []
nentry = 0
pos = None
if not args.gff:
args.qprint("No GFF used, translating sequences as pre-aligned in "
"coding frame.")
inputbuffer = []
current_contig = ''
for contigid, pos, allelesets in mvf.iterentries(decode=False):
if current_contig == '':
current_contig = contigid[:]
if contigid == current_contig:
inputbuffer.append((pos, allelesets))
else:
for _, amino_acids, alleles in iter_codons(
inputbuffer, mvf):
if all([x in '-X' for x in amino_acids]):
continue
if args.output_data == 'protein':
entrybuffer.append(
(current_contig, pos, (amino_acids,)))
else:
entrybuffer.append((
current_contig, pos, (
amino_acids, alleles[0],
alleles[1], alleles[2])))
nentry += 1
if nentry == args.line_buffer:
outmvf.write_entries(entrybuffer)
entrybuffer = []
nentry = 0
inputbuffer = [(pos, allelesets)]
current_contig = contigid[:]
if inputbuffer:
for _, amino_acids, alleles in iter_codons(
inputbuffer, outmvf):
if all([x in '-X' for x in amino_acids]):
continue
if args.output_data == 'protein':
entrybuffer.append(
(current_contig, pos, (amino_acids,)))
else:
entrybuffer.append((
current_contig, pos, (
amino_acids, alleles[0],
alleles[1], alleles[2])))
nentry += 1
if nentry == args.line_buffer:
outmvf.write_entries(entrybuffer)
entrybuffer = []
nentry = 0
else:
args.qprint("Indexing GFF gene names.")
# mvfid_to_gffname = outmvf.get_contig_reverse_dict()
for xcontig in outmvf.get_contig_indices():
mvf_entries = {}
xcontiglabel = outmvf.get_contig_labels(indices=xcontig)[0]
xcontigid = outmvf.get_contig_ids(indices=xcontig)[0]
if xcontiglabel not in gff:
if args.verbose:
print(
("No entries in GFF, "
"skipping contig: index:{} id:{} label:{}").format(
xcontig, xcontigid, xcontiglabel))
continue
if not xcontig % 100:
args.qprint("Processing contig: {} {}".format(
xcontigid, xcontiglabel))
for contigid, pos, allelesets in mvf.itercontigentries(
xcontig, decode=False):
mvf_entries[pos] = allelesets[0]
for coords in sorted(gff[xcontiglabel]):
reverse_strand = coords[3] == '-'
alleles = (tuple(mvf_entries.get(x, '-')
for x in coords[2::-1])
if reverse_strand is True
else tuple(mvf_entries.get(x, '-')
for x in coords[0:3]))
if all(len(x) == 1 for x in alleles):
if reverse_strand:
alleles = tuple(
MLIB.complement_bases[x] for x in alleles)
decoded_alleles = alleles
amino_acids = translate_single_codon(''.join(alleles))
else:
if reverse_strand is True:
decoded_alleles = tuple(tuple(MLIB.complement_bases[y]
for y in mvf.decode(x))
for x in alleles)
alleles = tuple(outmvf.encode(''.join(x))
for x in decoded_alleles)
else:
decoded_alleles = tuple(mvf.decode(x) for x in alleles)
amino_acids = tuple(translate_single_codon(''.join(x))
for x in zip(*decoded_alleles))
# print("aminx", amino_acids)
amino_acids = outmvf.encode(''.join(amino_acids))
# if all(x in '-X' for x in amino_acids):
# continue
# print("amino", amino_acids)
# print("translated", amino_acids, alleles)
if args.output_data == 'protein':
entrybuffer.append((xcontig, coords[0], (amino_acids,)))
else:
entrybuffer.append((
xcontigid, coords[0], (
amino_acids, alleles[0], alleles[1], alleles[2])))
nentry += 1
if nentry >= args.line_buffer:
args.qprint("Writing a block of {} entries.".format(
args.line_buffer))
outmvf.write_entries(entrybuffer)
entrybuffer = []
nentry = 0
if entrybuffer:
outmvf.write_entries(entrybuffer)
entrybuffer = []
nentry = 0
return ''
def infer_window_tree(args):
"""Main method"""
# ESTABLISH FILE OBJECTS
mvf = MultiVariantFile(args.mvf, 'read')
# Set up contig ids
if args.contig_ids is not None:
contig_ids = args.contig_ids[0].split(",")
elif args.contig_labels is not None:
contig_ids = mvf.get_contig_ids(
labels=args.contig_labels[0].split(","))
else:
contig_ids = mvf.get_contig_ids()
treefile = OutputFile(
args.out,
headers=[
'contig',
'windowstart',
'windowsize',
'tree',
'topology',
'topoid',
# 'templabels', ### USED FOR DEBUGGING ###
'alignlength',
'aligndepth',
'status'
])
topofile = OutputFile(args.out + '.counts',
headers=['rank', 'topology', 'count'])
if args.sample_indices is not None:
sample_indices = [int(x) for x in args.sample_indices[0].split(",")]
elif args.sample_labels is not None:
sample_indices = mvf.get_sample_indices(
labels=args.sample_labels[0].split(","))
else:
sample_indices = mvf.get_sample_indices()
if not os.path.exists(args.temp_dir):
os.mkdir(args.temp_dir)
os.chdir(args.temp_dir)
# SETUP PARAMS
main_labels = mvf.get_sample_labels(sample_indices)
if args.choose_allele in ['randomboth', 'majorminor']:
main_labels = [label + x for x in ['a', 'b'] for label in main_labels]
params = {
'outgroups':
args.raxml_outgroups or [],
'rootwith':
(args.root_with.split(',') if args.root_with is not None else None),
'minsites':
args.min_sites,
'minseqcoverage':
args.min_seq_coverage,
'mindepth':
args.min_depth,
'raxmlpath':
args.raxml_path,
'raxmlopts':
args.raxml_opts,
'duplicateseq':
args.duplicate_seq,
'model':
args.raxml_model,
'bootstrap':
args.bootstrap,
'windowsize':
args.windowsize,
'chooseallele':
args.choose_allele,
'tempdir':
args.temp_dir,
'tempprefix':
args.temp_prefix
}
# WINDOW START INTERATION
verify_raxml(params)
current_contig = ''
current_position = 0
window_data = None
skip_contig = False
topo_ids = {}
topo_counts = {}
for contig, pos, allelesets in mvf.iterentries(contigs=contig_ids,
subset=sample_indices,
quiet=args.quiet,
no_invariant=False,
no_ambig=False,
no_gap=False,
decode=True):
if current_contig == contig:
if skip_contig is True:
continue
if not same_window((current_contig, current_position),
(contig, pos), args.windowsize):
skip_contig = False
if window_data is not None:
entry = window_data.maketree_raxml(params)
if entry['status'] != 'ok':
if args.output_empty:
treefile.write_entry(entry)
if args.windowsize != -1:
skip_contig = True
else:
topo = entry["topology"]
topo_counts[topo] = topo_counts.get(topo, 0) + 1
if topo not in topo_ids:
topo_ids[topo] = (topo_ids
and max(topo_ids.values()) + 1 or 0)
entry["topoid"] = topo_ids[topo]
treefile.write_entry(entry)
current_position = (current_position + args.windowsize if
(contig == current_contig
and args.windowsize > 0) else 0)
current_contig = contig[:]
window_data = None
window_data = WindowData(
window_params={
'contigname': (mvf.get_contig_labels(
ids=current_contig) if args.output_contig_labels
is not None else current_contig[:]),
"windowstart": (
'-1' if args.windowsize == -1 else current_position +
0),
"windowsize":
args.windowsize,
"labels":
main_labels[:]
})
# ADD ALLELES
if mvf.flavor == 'dna':
if args.choose_allele != 'none':
allelesets[0] = hapsplit(allelesets[0], args.choose_allele)
window_data.append_alleles(allelesets[0], mindepth=args.min_depth)
# LAST LOOP
if window_data:
entry = window_data.maketree_raxml(params)
if entry['status'] != 'ok':
if args.output_empty:
treefile.write_entry(entry)
else:
topo = entry["topology"]
topo_counts[topo] = topo_counts.get(topo, 0) + 1
if topo not in topo_ids:
topo_ids[topo] = (max(topo_ids.values()) +
1 if topo_ids else 0)
entry["topoid"] = topo_ids[topo]
treefile.write_entry(entry)
window_data = None
# END WINDOW ITERATION
topo_list = sorted([(v, k) for k, v in topo_counts.items()], reverse=True)
for rank, [value, topo] in enumerate(topo_list):
topofile.write_entry({'rank': rank, 'count': value, 'topology': topo})
return ''
def infer_window_tree(args):
"""Main method"""
args.qprint("Running InferTree")
# ESTABLISH FILE OBJECTS
mvf = MultiVariantFile(args.mvf, 'read')
args.qprint("Read MVF File: {}".format(args.mvf))
# Set up contig ids
if args.contig_ids is not None:
contig_ids = args.contig_ids[0].split(",")
elif args.contig_labels is not None:
contig_ids = mvf.get_contig_ids(
labels=args.contig_labels[0].split(","))
else:
contig_ids = mvf.get_contig_ids()
treefile = OutputFile(
args.out,
headers=['contig', 'windowstart', 'windowsize', 'tree',
'topology', 'topoid',
# 'templabels', ### USED FOR DEBUGGING ###
'alignlength', 'aligndepth', 'status'])
topofile = OutputFile(args.out + '.counts',
headers=['rank', 'topology', 'count'])
if args.sample_indices is not None:
sample_indices = [int(x) for x in
args.sample_indices[0].split(",")]
elif args.sample_labels is not None:
sample_indices = mvf.get_sample_indices(
ids=args.sample_labels[0].split(","))
else:
sample_indices = mvf.get_sample_indices()
if not os.path.exists(args.temp_dir):
os.mkdir(args.temp_dir)
os.chdir(args.temp_dir)
# SETUP PARAMS
main_labels = mvf.get_sample_ids(sample_indices)
if args.choose_allele in ['randomboth', 'majorminor']:
main_labels = [label + x for x in ['a', 'b'] for label in main_labels]
params = {
'bootstrap': args.bootstrap,
'chooseallele': args.choose_allele,
'collapse_polytomies': args.collapse_polytomies,
'duplicateseq': args.duplicate_seq,
'engine': args.engine,
'engine_path': args.engine_path,
'engine_opts': args.engine_opts,
'mindepth': args.min_depth,
'minseqcoverage': args.min_seq_coverage,
'minsites': args.min_sites,
'model': args.model,
'outgroups': (args.raxml_outgroups
if args.raxml_outgroups is not None
else None),
'rootwith': (args.root_with.split(',')
if args.root_with is not None
else []),
'tempdir': args.temp_dir,
'tempprefix': args.temp_prefix,
'windowsize': args.windowsize,
}
# DEFAULT MODEL
if params['model'] is None:
if params['engine'] == 'raxml':
params['model'] = 'GTRGAMMA'
elif params['engine'] == 'raxml-ng':
params['model'] = "GTR+G"
# WINDOW START INTERATION
verify_raxml(params)
args.qprint("RAxML Found.")
current_contig = None
current_position = 0
window_data = None
# skip_contig = False
topo_ids = {}
topo_counts = {}
args.qprint("Prcocessing Records")
windowsizename = "window size={}".format(args.windowsize)
if windowsizename == "window size=-1":
windowsizename = "whole contig"
elif windowsizename == "window size=0":
windowsizename = "whole genome"
window_data = WindowData(window_params={
'contigname': 'all',
"windowstart": 0,
"windowsize": 0,
"labels": main_labels[:]})
for contig, pos, allelesets in mvf.iterentries(
contig_ids=contig_ids, subset=sample_indices,
no_invariant=False, no_ambig=False, no_gap=False, decode=True):
# if current_contig == contig:
# if skip_contig is True:
# args.qprint("Skipping contig: {}".format(current_contig))
# continue
if not same_window((current_contig, current_position),
(contig, pos), args.windowsize):
# skip_contig = False
if window_data is not None:
args.qprint(("Making tree for {} "
"at contig {} position {}").format(
windowsizename,
current_contig,
current_position))
entry = window_data.maketree_raxml(params)
if entry['status'] != 'ok':
if args.output_empty:
treefile.write_entry(entry)
# if args.windowsize != -1:
# skip_contig = True
args.qprint(
"TREE REJECTED with error code: {} ({})".format(
entry['status'], entry.get('comment', "None")))
else:
args.qprint("Tree completed.")
topo = entry["topology"]
topo_counts[topo] = topo_counts.get(topo, 0) + 1
if topo not in topo_ids:
topo_ids[topo] = (max(topo_ids.values()) + 1
if topo_ids else 0)
entry["topoid"] = topo_ids[topo]
treefile.write_entry(entry)
current_position = current_position + args.windowsize if (
contig == current_contig and args.windowsize > 0) else 0
current_contig = contig[:]
window_data = None
window_data = WindowData(window_params={
'contigname': (mvf.get_contig_labels(ids=current_contig) if
args.output_contig_labels is not None else
current_contig[:]),
"windowstart": ('-1' if args.windowsize == -1
else current_position + 0),
"windowsize": args.windowsize,
"labels": main_labels[:]})
# ADD ALLELES
if mvf.flavor == 'dna':
if args.choose_allele != 'none':
allelesets[0] = hapsplit(allelesets[0], args.choose_allele)
window_data.append_alleles(allelesets[0], mindepth=args.min_depth)
elif mvf.flavor == 'codon':
for i in (1, 2, 3):
if args.choose_allele != 'none':
allelesets[i] = hapsplit(allelesets[i], args.choose_allele)
window_data.append_alleles(allelesets[i], mindepth=args.min_depth)
# LAST LOOP
if window_data:
entry = window_data.maketree_raxml(params)
if entry['status'] != 'ok':
if args.output_empty:
treefile.write_entry(entry)
else:
topo = entry["topology"]
topo_counts[topo] = topo_counts.get(topo, 0) + 1
if topo not in topo_ids:
topo_ids[topo] = (
max(topo_ids.values()) + 1 if topo_ids else 0)
entry["topoid"] = topo_ids[topo]
treefile.write_entry(entry)
window_data = None
# END WINDOW ITERATION
topo_list = sorted([(v, k) for k, v in topo_counts.items()],
reverse=True)
for rank, [value, topo] in enumerate(topo_list):
topofile.write_entry({'rank': rank, 'count': value, 'topology': topo})
return ''
def translate_mvf(args):
"""Main method"""
mvf = MultiVariantFile(args.mvf, 'read')
if mvf.flavor != 'dna':
raise RuntimeError("MVF must be flavor=dna to translate")
if args.gff:
gff = parse_gff_translate(args.gff, args)
if not args.quiet:
print("gff_processed")
outmvf = MultiVariantFile(args.out, 'write', overwrite=args.overwrite)
outmvf.metadata = deepcopy(mvf.metadata)
outmvf.flavor = args.output_data
outmvf.write_data(outmvf.get_header())
entrybuffer = []
nentry = 0
if not args.gff:
inputbuffer = []
current_contig = ''
for contigid, pos, allelesets in mvf.iterentries(decode=False):
if current_contig == '':
current_contig = contigid[:]
if contigid == current_contig:
inputbuffer.append((pos, allelesets))
else:
for _, amino_acids, alleles in iter_codons(inputbuffer, mvf):
if all([x in '-X' for x in amino_acids]):
continue
if args.output_data == 'protein':
entrybuffer.append(
(current_contig, pos, (amino_acids, )))
else:
entrybuffer.append(
(current_contig, pos, (amino_acids, alleles[0],
alleles[1], alleles[2])))
nentry += 1
if nentry == args.line_buffer:
outmvf.write_entries(entrybuffer)
entrybuffer = []
nentry = 0
inputbuffer = [(pos, allelesets)]
current_contig = contigid[:]
if inputbuffer:
for _, amino_acids, alleles in iter_codons(inputbuffer, mvf):
if all([x in '-X' for x in amino_acids]):
continue
if args.output_data == 'protein':
entrybuffer.append((current_contig, pos, (amino_acids, )))
else:
entrybuffer.append(
(current_contig, pos, (amino_acids, alleles[0],
alleles[1], alleles[2])))
nentry += 1
if nentry == args.line_buffer:
outmvf.write_entries(entrybuffer)
entrybuffer = []
nentry = 0
else:
mvf_entries = {}
for contigid, pos, allelesets in mvf.iterentries(decode=False):
if contigid not in mvf_entries:
mvf_entries[contigid] = {}
mvf_entries[contigid][pos] = allelesets[0]
for contigname in sorted(gff):
contigid = mvf.get_contig_ids(labels=contigname)[0]
for coords in sorted(gff[contigname]):
reverse_strand = False
if coords[3] == '-':
reverse_strand = True
alleles = [
mvf_entries[contigid].get(x, '-')
for x in coords[2::-1]
]
else:
alleles = [
mvf_entries[contigid].get(x, '-') for x in coords[0:3]
]
if all(len(x) == 1 for x in alleles):
if reverse_strand:
alleles = [MLIB.complement_bases[x] for x in alleles]
decoded_alleles = alleles
amino_acids = translate(''.join(alleles))[0]
else:
if reverse_strand:
decoded_alleles = [[
MLIB.complement_bases[y] for y in mvf.decode(x)
] for x in alleles]
alleles = [
mvf.encode(''.join(x)) for x in decoded_alleles
]
else:
decoded_alleles = [mvf.decode(x) for x in alleles]
amino_acids = [
translate(''.join(x)) for x in zip(*decoded_alleles)
]
amino_acids = mvf.encode(''.join(
[x[0] for x in amino_acids]))
if all([x in '-X' for x in amino_acids]):
continue
if args.output_data == 'protein':
entrybuffer.append((contigid, coords[0], (amino_acids, )))
else:
entrybuffer.append(
(contigid, coords[0], (amino_acids, alleles[0],
alleles[1], alleles[2])))
nentry += 1
if nentry == args.line_buffer:
outmvf.write_entries(entrybuffer)
entrybuffer = []
nentry = 0
if entrybuffer:
outmvf.write_entries(entrybuffer)
entrybuffer = []
nentry = 0
return ''
def vcf2mvf(args=None):
"""Main method for vcf2mvf"""
sepchars = dict([("TAB", "\t"), ("SPACE", " "), ("DBLSPACE", " "),
("COMMA", ","), ("MIXED", None)])
args.fieldsep = sepchars[args.field_sep]
# ESTABLISH VCF
vcf = VariantCallFile(args.vcf, indexcontigs=(not args.no_autoindex))
# ESTABLISH MVF
mvf = MultiVariantFile(args.out, 'write', overwrite=args.overwrite)
# PROCESS CONTIG INFO
vcfcontigs = vcf.metadata['contigs'].copy()
contig_translate = {}
if args.contig_ids:
for cid, cvcf, cmvf in (x.split(';') for x in args.contig_ids):
try:
cid = int(cid)
except ValueError:
pass
assert cvcf in [vcfcontigs[x]['label'] for x in vcfcontigs]
for vid in vcfcontigs:
if vcfcontigs[vid]['label'] == cvcf:
contig_translate[cvcf] = [cid, cmvf]
if cid in mvf.metadata['contigs']:
raise RuntimeError(
'Contig id {} is not unique'.format(cid))
mvf.metadata['contigs'][cid] = vcfcontigs[vid].copy()
if cmvf in mvf.get_contig_labels():
raise RuntimeError(
'Contig label {} is not unique'.format(cmvf))
mvf.metadata['contigs'][cid]['label'] = cmvf[:]
mvf.reset_max_contig_id()
for vcid in vcfcontigs:
vlabel = vcfcontigs[vcid]['label']
if vlabel not in mvf.get_contig_labels():
if ((is_int(vlabel) or len(vlabel) < 3)
and vlabel not in mvf.get_contig_ids()):
newid = vlabel[:]
else:
newid = mvf.get_next_contig_id()
mvf.metadata['contigs'][newid] = vcfcontigs[vcid].copy()
contig_translate[vlabel] = [newid, vlabel]
mvf.reset_max_contig_id()
new_contigs = [(x, mvf.metadata['contigs'][x]['label'])
for x in mvf.metadata['contigs']]
for i, (newid, newlabel) in enumerate(new_contigs):
for j, (xid, xlabel) in enumerate(new_contigs):
if i == j:
continue
if newid == xlabel:
raise RuntimeError("Error contig id {} is the same as"
" the label for another contig"
" ({} {})".format(newid, xid, xlabel))
if newlabel == xid:
raise RuntimeError("Error contig label {} is the same"
"as the id for another contig"
"({} {})".format(newlabel, xid, xlabel))
# PROCESS SAMPLE INFO
samplelabels = [args.ref_label] + vcf.metadata['samples'][:]
if args.alleles_from:
args.alleles_from = args.alleles_from.split(':')
samplelabels += args.alleles_from
if args.sample_replace:
newsample = [
x.split(':') if ':' in tuple(x) else tuple([x, x])
for x in args.sample_replace
]
unmatched = [x for x in enumerate(samplelabels)]
for old, new in newsample:
labelmatched = False
for j, (i, name) in enumerate(unmatched):
if old in name:
samplelabels[i] = new
labelmatched = j
break
if labelmatched is not False:
del unmatched[labelmatched]
mvf.metadata['labels'] = samplelabels[:]
for i, label in enumerate(samplelabels):
mvf.metadata['samples'][i] = {'label': label}
mvf.metadata['ncol'] = len(mvf.metadata['labels'])
mvf.metadata['sourceformat'] = vcf.metadata['sourceformat']
# WRITE MVF HEADER
mvf.write_data(mvf.get_header())
mvfentries = []
nentry = 0
for vcfrecord in vcf.iterentries(args):
# try:
mvf_alleles = encode_mvfstring(''.join(vcfrecord['genotypes']))
if args.out_flavor in ('dnaqual', ):
qual_alleles = encode_mvfstring(''.join(vcfrecord['qscores']))
if mvf_alleles:
mvfentries.append(
(contig_translate.get(vcfrecord['contig'])[0],
vcfrecord['coord'],
((mvf_alleles,
qual_alleles) if args.out_flavor in ('dnaqual', ) else
(mvf_alleles, ))))
nentry += 1
if nentry == args.line_buffer:
mvf.write_entries(mvfentries, encoded=True)
mvfentries = []
nentry = 0
# except Exception as exception:
if mvfentries:
mvf.write_entries(mvfentries)
mvfentries = []
return ''