def _build_coverage(config, makefile, target, make_summary):
merged_nodes = []
coverage = _build_coverage_nodes(target)
for prefix in target.prefixes:
for (aoi_name, aoi_filename) in _get_aoi(prefix):
label = _get_prefix_label(prefix.label, aoi_name)
postfix = prefix.name if (not aoi_name) else (
"%s.%s" % (prefix.name, aoi_name))
files_and_nodes = _aggregate_for_prefix(coverage["Libraries"],
label)
output_filename = os.path.join(
config.destination, "%s.%s.coverage" % (target.name, postfix))
merged = MergeCoverageNode(input_files=files_and_nodes.keys(),
output_file=output_filename,
dependencies=files_and_nodes.values())
merged_nodes.append(merged)
description = "Libraries"
files_and_nodes = _aggregate_for_prefix(coverage["Libraries"], None)
if make_summary:
description = "Lanes and libraries"
files_and_nodes = _aggregate_for_prefix(coverage["Lanes"],
None,
into=files_and_nodes)
partial_nodes = MetaNode(description=description,
subnodes=files_and_nodes.values())
final_nodes = MetaNode(description="Final coverage", subnodes=merged_nodes)
coverage["Node"] = MetaNode(description="Coverage",
subnodes=(partial_nodes, final_nodes))
return coverage
def __init__(self, config, prefix, samples, features, target):
self.name = prefix["Name"]
self.label = prefix.get("Label") or self.name
self.reference = prefix["Reference"]
self.aoi = prefix.get("AreasOfInterest", {})
self.samples = safe_coerce_to_tuple(samples)
self.bams = {}
self.folder = config.destination
self.target = target
files_and_nodes = {}
for sample in self.samples:
files_and_nodes.update(sample.bams.iteritems())
if "Raw BAM" in features:
self.bams.update(
self._build_raw_bam(config, prefix, files_and_nodes))
if "Realigned BAM" in features:
self.bams.update(
self._build_realigned_bam(config, prefix, files_and_nodes))
sample_nodes = [sample.node for sample in self.samples]
if not self.bams:
for sample in self.samples:
self.bams.update(sample.bams)
self.node = MetaNode(description="Prefix: %s" % prefix["Name"],
dependencies=sample_nodes)
else:
self.node = MetaNode(description="Final BAMs: %s" % prefix["Name"],
subnodes=self.bams.values(),
dependencies=sample_nodes)
def __init__(self, config, prefix, node, log_file=None):
input_file, has_index = self._get_input_file(node)
subnodes, dependencies = [node], node.dependencies
if not has_index:
node = BAMIndexNode(infile=input_file, dependencies=node)
subnodes.append(node)
validation_params = ValidateBAMNode.customize(config=config,
input_bam=input_file,
output_log=log_file,
dependencies=node)
# Check MD tags against reference sequence
# FIXME: Disabled due to issues with Picard/Samtools disagreeing, backwards compatibility.
# validation_params.command.set_kwargs(IN_REFERENCE = prefix["Reference"])
# validation_params.command.add_option("R", "%(IN_REFERENCE)s", sep = "=")
# Ignored since we filter out misses and low-quality hits during mapping, which
# leads to a large proportion of missing mates for PE reads.
validation_params.command.add_option("IGNORE",
"MATE_NOT_FOUND",
sep="=")
# Ignored due to high rate of false positives for lanes with few hits, where
# high-quality reads may case ValidateSamFile to mis-identify the qualities
validation_params.command.add_option("IGNORE",
"INVALID_QUALITY_FORMAT",
sep="=")
subnodes.append(validation_params.build_node())
description = "<w/Validation: " + str(subnodes[0])[1:]
MetaNode.__init__(self,
description=description,
subnodes=subnodes,
dependencies=dependencies)
def build_phylogeny_nodes(options, settings, filtering, dependencies):
nodes = []
for (run_name, run_dd) in settings["PhylogeneticInference"].iteritems():
destination = os.path.join(options.destination, "phylogenies",
run_name)
if run_dd["PerGeneTrees"]:
run_nodes = []
for roi in run_dd["RegionsOfInterest"].itervalues():
roi_destination = os.path.join(destination, roi["Name"])
run_nodes.extend(
_build_examl_per_gene_nodes(options, settings, run_dd, roi,
roi_destination, filtering,
dependencies))
nodes.append(
MetaNode(description=run_name,
subnodes=run_nodes,
dependencies=dependencies))
else:
nodes.extend(
_build_examl_regions_nodes(options, settings, run_dd,
destination, filtering,
dependencies))
return MetaNode("Phylogenetic Inference", dependencies=nodes)
def __init__(self, config, reference, infiles, outfile, intervals=None,
dependencies=()):
if not intervals:
intervals = outfile + ".intervals"
infiles = safe_coerce_to_tuple(infiles)
trainer = _IndelTrainerNode(config=config,
reference=reference,
infiles=infiles,
outfile=intervals,
dependencies=dependencies)
aligner = _IndelRealignerNode(config=config,
reference=reference,
intervals=intervals,
infiles=infiles,
outfile=outfile,
dependencies=trainer)
description = "<GATK Indel Realigner: %i files -> '%s'>" \
% (len(infiles), outfile)
MetaNode.__init__(self,
description=description,
subnodes=[trainer, aligner],
dependencies=dependencies)
def __init__(self, config, prefix, node, log_file=None):
input_file, has_index = self._get_input_file(node)
subnodes, dependencies = [node], node.dependencies
if not has_index:
node = BAMIndexNode(infile=input_file,
dependencies=node)
subnodes.append(node)
validation_params = ValidateBAMNode.customize(config=config,
input_bam=input_file,
output_log=log_file,
dependencies=node)
# Check MD tags against reference sequence
# FIXME: Disabled due to issues with Picard/Samtools disagreeing,
# backwards compatibility. See the discussion at
# http://sourceforge.net/mailarchive/message.php?msg_id=31348639
# validation_params.command.set_kwargs(IN_REF=prefix["Reference"])
# validation_params.command.add_option("R", "%(IN_REF)s", sep="=")
# Ignored since we may filter out misses and low-quality hits during
# mapping, which leads to a large proportion of missing PE mates.
validation_params.command.add_option("IGNORE", "MATE_NOT_FOUND",
sep="=")
# Ignored due to high rate of false positives for lanes with few hits,
# where high-quality reads may cause mis-identification of qualities
validation_params.command.add_option("IGNORE",
"INVALID_QUALITY_FORMAT", sep="=")
subnodes.append(validation_params.build_node())
description = "<w/Validation: " + str(subnodes[0])[1:]
MetaNode.__init__(self,
description=description,
subnodes=subnodes,
dependencies=dependencies)
def build_codeml_nodes(options, settings, interval, taxa, filtering,
dependencies):
in_postfix, out_postfix, afa_ext = "", "", ".afa"
if any(filtering.itervalues()):
in_postfix = out_postfix = ".filtered"
if not settings["MSAlignment"]["Enabled"]:
out_postfix = ".unaligned" + out_postfix
afa_ext = ".fasta"
paml = settings["PAML"]
sequences = common.collect_sequences(options, interval, taxa)
sequencedir = os.path.join(options.destination, "alignments",
interval["Name"] + in_postfix)
destination = os.path.join(options.destination, "paml", "codeml",
interval["Name"] + out_postfix)
# Build meta-node for sequence conversion to PHYLIP format accepted by codeml
phylip_nodes = {}
for sequence in sequences:
input_file = os.path.join(sequencedir, sequence + afa_ext)
output_file = os.path.join(destination, sequence + ".phy")
phylip_nodes[sequence] = FastaToPAMLPhyNode(
input_file=input_file,
output_file=output_file,
exclude_groups=paml["codeml"]["ExcludeGroups"],
dependencies=dependencies)
phylip_meta = MetaNode(description = "<FastaToPAMLPhyNodes: '%s/*.%s' -> '%s/*.phy'>" \
% (sequencedir, afa_ext, destination),
subnodes = phylip_nodes.values(),
dependencies = dependencies)
codeml_nodes = []
for (ctl_name, ctl_file) in paml["codeml"]["Control Files"].iteritems():
for (sequence, node) in phylip_nodes.iteritems():
output_prefix = os.path.join(destination,
sequence + ".%s" % (ctl_name, ))
codeml = CodemlNode(control_file=ctl_file,
trees_file=paml["codeml"]["Tree File"],
sequence_file=iter(node.output_files).next(),
output_prefix=output_prefix,
dependencies=node)
codeml_nodes.append(codeml)
return MetaNode(description="<CodemlNodes>",
subnodes=codeml_nodes,
dependencies=phylip_meta)
def __init__(self, input_files, destination, filter_by, dependencies=()):
subnodes = []
filter_by = dict(filter_by)
for (filename, node) in input_files.iteritems():
output_filename = fileutils.reroot_path(destination, filename)
subnodes.append(FilterSingletonsNode(input_file=filename,
output_file=output_filename,
filter_by=filter_by,
dependencies=node))
MetaNode.__init__(self,
description="<FilterSingleton: %i files -> '%s'>"
% (len(subnodes), destination),
subnodes=subnodes,
dependencies=dependencies)
def __init__(self, input_files, destination, filter_by, dependencies=()):
subnodes = []
filter_by = dict(filter_by)
for (filename, node) in input_files.iteritems():
output_filename = fileutils.reroot_path(destination, filename)
subnodes.append(FilterSingletonsNode(input_file=filename,
output_file=output_filename,
filter_by=filter_by,
dependencies=node))
MetaNode.__init__(self,
description="<FilterSingleton: %i files -> '%s'>"
% (len(subnodes), destination),
subnodes=subnodes,
dependencies=dependencies)
def __init__(self, config, prefixes, name):
self.name = name
self.prefixes = safe_coerce_to_tuple(prefixes)
self._nodes_alignment = MetaNode(description = "Alignments:",
dependencies = [prefix.node for prefix in self.prefixes])
self._nodes_extras = {}
def __init__(self, rootdir, sequences, preset = "auto", subnodes = (), dependencies = ()):
subnodes = []
for sequence in sequences:
prefix = os.path.join(rootdir, sequence)
node = MAFFTNode(infile = prefix + ".fasta",
outfile = prefix + ".afa",
preset = preset,
dependencies = dependencies)
subnodes.append(node)
MetaNode.__init__(self,
description = "<MAFFTAlignSequences (%s): In '%s'>" \
% (preset, rootdir),
subnodes = subnodes,
dependencies = dependencies)
def build_taxa_nodes(options, genotyping, intervals, taxa, dependencies=()):
nodes = []
for interval in intervals.itervalues():
interval = deepcopy(interval)
# Override default genome (BAM file) if specified
interval["Genome"] = common.get_genome_for_interval(interval, taxa)
# Enforce homozygous contigs based on gender tag
# interval[
interval["Homozygous Contigs"] = interval["Homozygous Contigs"][
taxa["Gender"]]
genotyping_method = taxa.get("Genotyping Method", "samtools").lower()
if genotyping_method == "reference sequence":
nodes.extend(
build_reference_nodes(options, taxa, interval, dependencies))
elif genotyping_method == "random sampling":
nodes.extend(
build_sampling_nodes(options, genotyping, taxa, interval,
dependencies))
elif genotyping_method == "samtools":
nodes.extend(
build_genotyping_nodes(options, genotyping, taxa, interval,
dependencies))
return MetaNode(description=taxa["Name"], dependencies=nodes)
def _build_depth(config, target):
nodes = []
for prefix in target.prefixes:
for (roi_name, roi_filename) in _get_roi(prefix, name_prefix="."):
if roi_filename is not None:
# ROIs require indexed access, and hence that the final BAM
# (either raw or realigned) has been built. By default, the
# the realigned BAM is used (based on lexical order).
bam_files = tuple(sorted(prefix.bams.items()))
input_files, dependencies = bam_files[-1]
else:
input_files = {}
for sample in prefix.samples:
input_files.update(sample.bams)
dependencies = input_files.values()
input_files = input_files.keys()
output_filename = "%s.%s%s.depths" % (target.name, prefix.name,
roi_name)
output_fpath = os.path.join(config.destination, output_filename)
node = DepthHistogramNode(config=config,
target_name=target.name,
input_files=input_files,
regions_file=roi_filename,
output_file=output_fpath,
dependencies=dependencies)
nodes.append(node)
return MetaNode(description="DepthHistograms", subnodes=nodes)
def __init__(self, config, target, prefix, lanes, name):
self.name = name
self.lanes = safe_coerce_to_tuple(lanes)
self.options = lanes[0].options
self.folder = os.path.dirname(self.lanes[0].folder)
self.bams = None
self.mapdamage = None
assert all((self.folder == os.path.dirname(lane.folder))
for lane in self.lanes)
assert all((self.options == lane.options) for lane in self.lanes)
lane_bams = self._collect_bams_by_type(self.lanes)
self.datadup_check = self._build_dataduplication_node(lane_bams)
self.duphist = \
self._build_duphist_nodes(config, target, prefix, lane_bams)
pcr_duplicates = self.options["PCRDuplicates"]
if pcr_duplicates:
lane_bams = self._remove_pcr_duplicates(config, prefix, lane_bams,
pcr_duplicates)
# At this point we no longer need to differentiate between types of reads
files_and_nodes = self._collect_files_and_nodes(lane_bams)
self.bams, self.mapdamage = \
self._build_mapdamage_nodes(config, target, prefix, files_and_nodes)
self.node = MetaNode(
description="Library: %s" % os.path.basename(self.folder),
dependencies=self.bams.values() + [self.datadup_check])
def test_metanode__nodes():
subnodes = [Node(), Node()]
dependencies = [Node(), Node()]
node = MetaNode(subnodes = iter(subnodes),
dependencies = iter(dependencies))
assert_equal(node.subnodes, frozenset(subnodes))
assert_equal(node.dependencies, frozenset(dependencies))
def build_sample_nodes(options,
genotyping,
regions_sets,
sample,
dependencies=()):
nodes = []
for regions in regions_sets.itervalues():
regions = deepcopy(regions)
# Enforce homozygous contigs based on gender tag
regions["HomozygousContigs"] \
= regions["HomozygousContigs"][sample["Gender"]]
genotyping_method = sample["GenotypingMethod"].lower()
if genotyping_method not in _GENOTYPING_METHODS:
assert False, "Unexpected genotyping method %r for sample %r" \
% (genotyping_method, sample["Name"])
genotyping_function = _GENOTYPING_METHODS[genotyping_method]
node = genotyping_function(options=options,
genotyping=genotyping[regions["Name"]],
sample=sample["Name"],
regions=regions,
dependencies=dependencies)
nodes.extend(node)
return MetaNode(description=sample["Name"], dependencies=nodes)
def test_metanode__properties():
node = MetaNode()
assert_equal(node.input_files, frozenset())
assert_equal(node.output_files, frozenset())
assert_equal(node.executables, frozenset())
assert_equal(node.auxiliary_files, frozenset())
assert_equal(node.requirements, frozenset())
def test_bwa(config):
index = _bwa_index(config)
aln_se = _bwa_aln_se(config, index)
aln_pe = _bwa_aln_pe(config, index)
sw_se = _bwa_sw_se(config, index)
sw_pe = _bwa_sw_pe(config, index)
return MetaNode(description="BWA",
dependencies=(aln_se, aln_pe, sw_se, sw_pe))
def test_bwa(config):
index = _bwa_index(config)
dependencies = []
for algorithm in ("backtrack", "bwasw", "mem"):
dependencies.append(_bwa_se(config, index, algorithm))
dependencies.append(_bwa_pe(config, index, algorithm))
return MetaNode(description="BWA",
dependencies=dependencies)
def index_references(config, makefiles):
references = {}
references_bwa = {}
references_bowtie2 = {}
for makefile in makefiles:
for subdd in makefile["Prefixes"].itervalues():
reference = subdd["Reference"]
if reference not in references:
# Validation of the FASTA file; not blocking for the other
# steps, as it is only expected to fail very rarely, but will
# block subsequent analyses depending on the FASTA.
valid_node = ValidateFASTAFilesNode(input_files=reference,
output_file=reference +
".validated")
# Indexing of FASTA file using 'samtools faidx'
faidx_node = FastaIndexNode(reference)
# Indexing of FASTA file using 'BuildSequenceDictionary.jar'
dict_node = BuildSequenceDictNode(config=config,
reference=reference,
dependencies=(valid_node, ))
# Indexing of FASTA file using 'bwa index'
bwa_node = BWAIndexNode(input_file=reference,
dependencies=(valid_node, ))
# Indexing of FASTA file using ''
bowtie2_node = Bowtie2IndexNode(input_file=reference,
dependencies=(valid_node, ))
references[reference] = \
MetaNode(description="Reference Sequence",
dependencies=(valid_node, faidx_node, dict_node))
references_bwa[reference] = \
MetaNode(description="Reference Sequence",
dependencies=(valid_node, faidx_node,
dict_node, bwa_node))
references_bowtie2[reference] = \
MetaNode(description="Reference Sequence",
dependencies=(valid_node, faidx_node,
dict_node, bowtie2_node))
subdd["Node"] = references[reference]
subdd["Node:BWA"] = references_bwa[reference]
subdd["Node:Bowtie2"] = references_bowtie2[reference]
def __init__(self, config, prefixes, name):
self.name = name
self.prefixes = safe_coerce_to_tuple(prefixes)
self._nodes_extras = {}
self._nodes_alignment = MetaNode(description = "Alignments:",
dependencies = [prefix.node for prefix in self.prefixes])
self._setup_extra_nodes("mapDamage", "mapdamage")
self._setup_extra_nodes("Duplicate Histogram", "duphist")
def build_variant_nodes(options,reference, group, dependencies = ()):
gatk_outfile = os.path.join(
options.makefile['OutDir'],"gatk.{}.{}.raw.vcf".format(
group['Group'],reference['Label'])
)
# Build the GATK Variant Calling Node
gatk_variants = UnifiedGenotyperNode.customize(
reference = reference['Path'],
infiles = [ os.path.join(options.makefile['BaseDir'],
ind,
ind + "."+ reference['Label'] + ".realigned.bam")
for ind in group['Inds']],
outfile = gatk_outfile,
options = options
)
# Build the SAMTOOLs Variant Calling Node
samtools_outfile = os.path.join(options.makefile['OutDir'],
"samtools.{}.{}.raw.vcf".format(
group['Group'],reference['Label'])
)
samtools_variants = VariantNode.customize(
reference = reference['Path'],
infiles = [ os.path.join(options.makefile['BaseDir'],
ind,
ind + "."+ reference['Label'] + ".realigned.bam")
for ind in group['Inds']],
outfile = samtools_outfile,
options = options
)
samtools_variants = samtools_variants.build_node()
gatk_variants = gatk_variants.build_node()
# Build the Variant Filtering Nodes
intersect_variants = VariantFilterNode.customize(
reference = reference['Path'],
infile = gatk_outfile,
outfile = os.path.join(
gatk_outfile.replace(".raw.vcf",".gatk_samtools_intersect.vcf")
),
filters = {
"--intersect_vcf" : gatk_outfile.replace(
options.makefile['intersect_vcf']['replace'],
options.makefile['intersect_vcf']['with']
),
"--emit": "pass"
},
options = options,
dependencies = [gatk_variants,samtools_variants]
)
intersect_variants = intersect_variants.build_node()
return MetaNode(description = "Variant Intersection",
dependencies = [intersect_variants]
)
def __init__(self, config, reference, infiles, outfile, intervals=None, dependencies=()):
if not intervals:
intervals = outfile + ".intervals"
infiles = safe_coerce_to_tuple(infiles)
trainer = _IndelTrainerNode(
config=config, reference=reference, infiles=infiles, outfile=intervals, dependencies=dependencies
)
aligner = _IndelRealignerNode(
config=config,
reference=reference,
intervals=intervals,
infiles=infiles,
outfile=outfile,
dependencies=trainer,
)
description = "<GATK Indel Realigner: %i files -> '%s'>" % (len(infiles), outfile)
MetaNode.__init__(self, description=description, subnodes=[trainer, aligner], dependencies=dependencies)
def index_references(config, makefiles):
references = {}
for makefile in makefiles:
for dd in makefile["Prefixes"].itervalues():
reference = os.path.realpath(dd["Reference"])
if reference not in references:
references[reference] = \
MetaNode(description = "Reference Sequence",
dependencies = (FastaIndexNode(dd["Reference"]),
BuildSequenceDictNode(config = config,
reference = dd["Reference"])))
dd["Node"] = references[reference]
def __init__(self, config, prefix, libraries, name):
self.name = name
self.bams = {}
self.libraries = safe_coerce_to_tuple(libraries)
for library in self.libraries:
self.bams.update(library.bams.iteritems())
self.folder = os.path.dirname(self.libraries[0].folder)
self.node = MetaNode(
description="Sample: %s" % os.path.basename(self.folder),
dependencies=[library.node for library in self.libraries])
def _adapterremoval_se(config):
node_params = {"input_files" : ("tests/data/raw_reads/se_reads_R1_001.fastq.gz",
"tests/data/raw_reads/se_reads_R1_002.fastq.gz"),
"dependencies" : config.dependencies}
standard = SE_AdapterRemovalNode(output_prefix = os.path.join(config.destination, "se_standard"),
**node_params)
custom = SE_AdapterRemovalNode.customize(output_prefix = os.path.join(config.destination, "se_custom"),
**node_params)
custom.command.set_option("--minlength", 30)
return MetaNode(description = "AdapterRemoval_SE",
dependencies = [standard, custom.build_node()])
def _setup_extra_nodes(self, name, key):
nodes = []
for prefix in self.prefixes:
prefix_nodes = []
for sample in prefix.samples:
for library in sample.libraries:
value = getattr(library, key)
if value:
prefix_nodes.append(value)
if any(prefix_nodes):
node = MetaNode(description=prefix.name,
subnodes=prefix_nodes)
nodes.append(node)
if nodes:
self.add_extra_nodes(name, nodes)
def _bwa_se(config, index, algorithm):
node_params = {"input_file_1": "tests/data/sim_reads/mate_1.fastq.gz",
"prefix": os.path.join(config.destination, "rCRS"),
"reference": "tests/data/rCRS.fasta",
"algorithm": algorithm,
"dependencies": (config.dependencies, index)}
template = os.path.join(config.destination, "%s_se_%s", "output.bam")
std_bam = template % (algorithm, "standard")
standard = BWANode(output_file=std_bam, **node_params).build_node()
cus_bam = template % (algorithm, "custom")
custom = BWANode(output_file=cus_bam, **node_params)
custom.commands["convert"].add_option("--rg-id", "myRG")
custom.commands["convert"].add_option("--rg", "PL:Illumina")
custom = custom.build_node()
return MetaNode(description="BWA %s SE" % algorithm,
dependencies=[standard, custom])
def _bwa_sw_se(config, index):
node_params = {
"input_file_1": "tests/data/sim_reads/mate_1.fastq.gz",
"prefix": os.path.join(config.destination, "rCRS"),
"reference": "tests/data/rCRS.fasta",
"dependencies": (config.dependencies, index)
}
standard = BWASWNode(output_file=os.path.join(config.destination,
"sw_se_standard",
"output.bam"),
**node_params)
custom = BWASWNode.customize(output_file=os.path.join(
config.destination, "sw_se_custom", "output.bam"),
**node_params)
custom.commands["aln"].set_option("-z", 10)
return MetaNode(description="BWA SW SE",
dependencies=[standard, custom.build_node()])
def _bowtie2_aln_pe(config, index):
node_params = {
"input_file_1": "tests/data/sim_reads/mate_1.fastq.gz",
"input_file_2": "tests/data/sim_reads/mate_2.fastq.gz",
"prefix": os.path.join(config.destination, "rCRS"),
"reference": "tests/data/rCRS.fasta",
"dependencies": (config.dependencies, index)
}
standard = Bowtie2Node(output_file=os.path.join(config.destination,
"aln_pe_standard",
"output.bam"),
**node_params)
custom = Bowtie2Node.customize(output_file=os.path.join(
config.destination, "aln_pe_custom", "output.bam"),
**node_params)
custom.commands["aln"].set_option("--very-sensitive")
return MetaNode(description="Bowtie2 aln PE",
dependencies=[standard, custom.build_node()])
def _add_mapdamage_nodes(config, makefile, target):
if "mapDamage" not in makefile["Options"]["Features"]:
return
nodes = []
for prefix in target.prefixes:
libraries = []
for sample in prefix.samples:
for library in sample.libraries:
folder = os.path.join(config.destination, "%s.%s.mapDamage" % (target.name, prefix.name), library.name)
libraries.append(MapDamageNode(config = config,
reference = prefix.reference,
input_files = library.bams.keys(),
output_directory = folder,
dependencies = library.bams.values()))
nodes.append(MetaNode(description = prefix.name,
subnodes = libraries))
target.add_extra_nodes("mapDamage", nodes)
def _bwa_aln_se(config, index):
node_params = {
"input_file": "tests/data/sim_reads/mate_1.fastq.gz",
"prefix": os.path.join(config.destination, "rCRS"),
"reference": "tests/data/rCRS.fasta",
"dependencies": (config.dependencies, index)
}
standard = SEBWANode(output_file=os.path.join(config.destination,
"aln_se_standard",
"output.bam"),
**node_params)
custom = SEBWANode.customize(output_file=os.path.join(
config.destination, "aln_se_custom", "output.bam"),
**node_params)
custom.commands["samse"].set_option(
"-r", "@RG\tID:1\tPL:Illumina\tPU:123456\tLB:Library_1\tSM:Sample_1")
return MetaNode(description="BWA aln SE",
dependencies=[standard, custom.build_node()])
def build_snp_list(groups,prefix,options,dependencies = ()):
recal_files = glob.glob(
os.path.join(options.makefile['RecalDir'],"gatk*group_only*.vcf")
)
VQSLOD_cutoff = float(options.makefile['VQSLOD_cutoff'])
recal_merge = VariantMergeNode.customize(
vcf_list = recal_files,
outfile = os.path.join(options.makefile['RecalDir'],
'RECAL_MERGED.{}.vcf'.format(prefix['Label'])
),
reference = prefix['Path'],
options = options,
dependencies = dependencies
)
recal_merge = recal_merge.build_node()
recal_filter = VariantFilterNode.customize(
reference = prefix['Path'],
infile = os.path.join(options.makefile['RecalDir'],
'RECAL_MERGED.{}.vcf'.format(prefix['Label'])
),
outfile = os.path.join(options.makefile['RecalDir'],
"RECAL_FILTERED.{}.vcf".format(prefix['Label'])
),
filters = {
"--not_within" : '20',
"--allelic" : '2',
"--repetitive" : "Repeat_Regions.txt",
"--emit" : "pass",
"--field" : 'AC',
"--thresh" : '2'
},
options = options,
dependencies = dependencies + [recal_merge]
)
recal_filter = recal_filter.build_node()
return MetaNode(description="SNP List Node",
dependencies = [recal_filter]
)