def _perform_update(self, isa, ped_donors):
# Traverse investigation, studies, assays, potentially updating the nodes.
donor_map = self._build_donor_map(ped_donors)
visitor = SheetUpdateVisitor(donor_map, self.config)
iwalker = isa_support.InvestigationTraversal(isa.investigation, isa.studies, isa.assays)
iwalker.run(visitor)
investigation, studies, assays = iwalker.build_evolved()
# Add records to study and assay for donors not seen so far.
todo_ped_donors = [
donor for donor in donor_map.values() if donor.name not in visitor.seen_source_names
]
studies, assays = isa_germline_append_donors(
studies, assays, tuple(todo_ped_donors), tuple(visitor.seen_sample_names), self.config
)
new_isa = attr.evolve(isa, investigation=investigation, studies=studies, assays=assays)
# Write ISA-tab into string buffers.
io_investigation = io.StringIO()
InvestigationWriter.from_stream(isa.investigation, io_investigation).write()
ios_studies = {}
for name, study in new_isa.studies.items():
ios_studies[name] = io.StringIO()
StudyWriter.from_stream(study, ios_studies[name]).write()
ios_assays = {}
for name, assay in new_isa.assays.items():
ios_assays[name] = io.StringIO()
AssayWriter.from_stream(assay, ios_assays[name]).write()
# Write out updated ISA-tab files using the diff helper.
i_path = pathlib.Path(self.config.input_investigation_file)
overwrite_helper(
i_path,
io_investigation.getvalue(),
do_write=not self.config.dry_run,
show_diff=True,
show_diff_side_by_side=self.config.show_diff_side_by_side,
answer_yes=self.config.yes,
)
for filename, ios_study in ios_studies.items():
overwrite_helper(
i_path.parent / filename,
ios_study.getvalue(),
do_write=not self.config.dry_run,
show_diff=True,
show_diff_side_by_side=self.config.show_diff_side_by_side,
answer_yes=self.config.yes,
)
for filename, ios_assay in ios_assays.items():
overwrite_helper(
i_path.parent / filename,
ios_assay.getvalue(),
do_write=not self.config.dry_run,
show_diff=True,
show_diff_side_by_side=self.config.show_diff_side_by_side,
answer_yes=self.config.yes,
)
def _perform_update(self, isa, annotation_map, header_map):
# Traverse investigation, studies, assays, potentially updating the nodes.
visitor = SheetUpdateVisitor(
annotation_map,
header_map,
self.config.force_update,
self.config.target_study,
self.config.target_assay,
)
iwalker = isa_support.InvestigationTraversal(isa.investigation, isa.studies, isa.assays)
iwalker.run(visitor)
investigation, studies, assays = iwalker.build_evolved()
new_isa = attr.evolve(isa, investigation=investigation, studies=studies, assays=assays)
# Write ISA-tab into string buffers.
io_investigation = io.StringIO()
InvestigationWriter.from_stream(isa.investigation, io_investigation).write()
ios_studies = {}
for name, study in new_isa.studies.items():
ios_studies[name] = io.StringIO()
StudyWriter.from_stream(study, ios_studies[name]).write()
ios_assays = {}
for name, assay in new_isa.assays.items():
ios_assays[name] = io.StringIO()
AssayWriter.from_stream(assay, ios_assays[name]).write()
# Write out updated ISA-tab files using the diff helper.
i_path = pathlib.Path(self.config.input_investigation_file)
overwrite_helper(
i_path,
io_investigation.getvalue(),
do_write=not self.config.dry_run,
show_diff=True,
show_diff_side_by_side=self.config.show_diff_side_by_side,
answer_yes=self.config.yes,
)
for filename, ios_study in ios_studies.items():
overwrite_helper(
i_path.parent / filename,
ios_study.getvalue(),
do_write=not self.config.dry_run,
show_diff=True,
show_diff_side_by_side=self.config.show_diff_side_by_side,
answer_yes=self.config.yes,
)
for filename, ios_assay in ios_assays.items():
overwrite_helper(
i_path.parent / filename,
ios_assay.getvalue(),
do_write=not self.config.dry_run,
show_diff=True,
show_diff_side_by_side=self.config.show_diff_side_by_side,
answer_yes=self.config.yes,
)
def _parse_write_assert_assay(investigation_file,
tmp_path,
quote=None,
normalize=False,
skip=None):
# Load investigation
investigation = InvestigationReader.from_stream(investigation_file).read()
InvestigationValidator(investigation).validate()
directory = os.path.normpath(os.path.dirname(investigation_file.name))
# Iterate assays
for s, study_info in enumerate(investigation.studies):
for a, assay_info in enumerate(study_info.assays):
if skip and str(assay_info.path) in skip:
continue
# Load assay
path_in = os.path.join(directory, assay_info.path)
with open(path_in, "rt") as inputf:
assay = AssayReader.from_stream("S{}".format(s + 1),
"A{}".format(a + 1),
inputf).read()
AssayValidator(investigation, study_info, assay_info,
assay).validate()
# Write assay to temporary file
path_out = tmp_path / assay_info.path
with open(path_out, "wt", newline="") as file:
AssayWriter.from_stream(assay, file, quote=quote).write()
if normalize:
# Read and write assay again
path_in = path_out
with open(path_out, "rt") as inputf:
assay = AssayReader.from_stream("S{}".format(s + 1),
"A{}".format(a + 1),
inputf).read()
AssayValidator(investigation, study_info, assay_info,
assay).validate()
path_out = tmp_path / (assay_info.path.name + "_b")
with open(path_out, "wt", newline="") as file:
AssayWriter.from_stream(assay, file, quote=quote).write()
# Sort and compare input and output
path_in_s = tmp_path / (assay_info.path.name + ".in.sorted")
path_out_s = tmp_path / (assay_info.path.name + ".out.sorted")
assert filecmp.cmp(sort_file(path_in, path_in_s),
sort_file(path_out, path_out_s),
shallow=False)
def run_writing(args, path_out, investigation, studies, assays):
# Write investigation
if args.output_investigation_file.name == "<stdout>":
InvestigationWriter.from_stream(investigation,
args.output_investigation_file,
quote=args.quotes).write()
else:
with open(args.output_investigation_file.name, "wt",
newline="") as outputf:
InvestigationWriter.from_stream(investigation,
outputf,
quote=args.quotes).write()
# Write studies and assays
for s, study_info in enumerate(investigation.studies):
if args.output_investigation_file.name == "<stdout>":
if study_info.info.path:
StudyWriter.from_stream(studies[s],
args.output_investigation_file,
quote=args.quotes).write()
for a, assay_info in enumerate(study_info.assays):
if assay_info.path:
AssayWriter.from_stream(assays[s][a],
args.output_investigation_file,
quote=args.quotes).write()
else:
if study_info.info.path:
with open(os.path.join(path_out, study_info.info.path),
"wt",
newline="") as outputf:
StudyWriter.from_stream(studies[s],
outputf,
quote=args.quotes).write()
for a, assay_info in enumerate(study_info.assays):
if assay_info.path:
with open(os.path.join(path_out, assay_info.path),
"wt",
newline="") as outputf:
AssayWriter.from_stream(assays[s][a],
outputf,
quote=args.quotes).write()
def create_and_write(out_path):
"""Create an investigation with a study and assay and write to ``output_path``."""
# Prepare one or more study sections
# Prepare basic study information
study_info = models.BasicInfo(
path="s_minimal.txt",
identifier="s_minimal",
title="Germline Study",
description=None,
submission_date=None,
public_release_date=None,
comments=(
models.Comment(name="Study Grant Number", value=None),
models.Comment(name="Study Funding Agency", value=None),
),
headers=[],
)
# Create one or more assays
assay_01 = models.AssayInfo(
measurement_type=models.OntologyTermRef(
name="exome sequencing assay",
accession="http://purl.obolibrary.org/obo/OBI_0002118",
ontology_name="OBI",
),
technology_type=models.OntologyTermRef(
name="nucleotide sequencing",
accession="http://purl.obolibrary.org/obo/OBI_0000626",
ontology_name="OBI",
),
platform=None,
path="a_minimal.txt",
comments=(),
headers=[],
)
# Prepare one or more protocols
protocol_01 = models.ProtocolInfo(
name="sample collection",
type=models.OntologyTermRef(name="sample collection"),
description=None,
uri=None,
version=None,
parameters={},
components={},
comments=(),
headers=[],
)
protocol_02 = models.ProtocolInfo(
name="nucleic acid sequencing",
type=models.OntologyTermRef(name="nucleic acid sequencing"),
description=None,
uri=None,
version=None,
parameters={},
components={},
comments=(),
headers=[],
)
# Create study object
study_01 = models.StudyInfo(
info=study_info,
designs=(),
publications=(),
factors={},
assays=(assay_01, ),
protocols={
protocol_01.name: protocol_01,
protocol_02.name: protocol_02
},
contacts=(),
)
# Prepare other investigation section
# Prepare one or more ontology term source references
onto_ref_01 = models.OntologyRef(
name="OBI",
file="http://data.bioontology.org/ontologies/OBI",
version="31",
description="Ontology for Biomedical Investigations",
comments=(),
headers=[],
)
# Prepare basic investigation information
invest_info = models.BasicInfo(
path="i_minimal.txt",
identifier="i_minimal",
title="Minimal Investigation",
description=None,
submission_date=None,
public_release_date=None,
comments=(),
headers=[],
)
# Create investigation object
investigation = models.InvestigationInfo(
ontology_source_refs={onto_ref_01.name: onto_ref_01},
info=invest_info,
publications=(),
contacts=(),
studies=(study_01, ),
)
# Validate investigation
InvestigationValidator(investigation).validate()
# Write the investigation as ISA-Tab txt file
with open(join(out_path, investigation.info.path), "wt",
newline="") as outputf:
InvestigationWriter.from_stream(investigation=investigation,
output_file=outputf).write()
# Create a corresponding Study graph
# Create at least on source, one sample and one collection process
# Unique names are required for unambiguous node identification
source_01 = models.Material(
type="Source Name",
unique_name="S1-source-0815",
name="0815",
extract_label=None,
characteristics=(),
comments=(),
factor_values=(),
material_type=None,
headers=[table_headers.SOURCE_NAME],
)
sample_01 = models.Material(
type="Sample Name",
unique_name="S1-sample-0815-N1",
name="0815-N1",
extract_label=None,
characteristics=(),
comments=(),
factor_values=(),
material_type=None,
headers=[table_headers.SAMPLE_NAME],
)
process_01 = models.Process(
protocol_ref="sample collection",
unique_name="S1-sample collection-2-1",
name=None,
name_type=None,
date=None,
performer=None,
parameter_values=(),
comments=(),
array_design_ref=None,
first_dimension=None,
second_dimension=None,
headers=[table_headers.PROTOCOL_REF],
)
# Create the arcs to connect the material and process nodes, referenced by the unique name
arc_01 = models.Arc(tail="S1-source-0815", head="S1-sample collection-2-1")
arc_02 = models.Arc(tail="S1-sample collection-2-1",
head="S1-sample-0815-N1")
# Create the study graph object
study_graph_01 = models.Study(
file=investigation.studies[0].info.path,
header=None,
materials={
source_01.unique_name: source_01,
sample_01.unique_name: sample_01
},
processes={process_01.unique_name: process_01},
arcs=(arc_01, arc_02),
)
# Validate study graph
StudyValidator(investigation=investigation,
study_info=investigation.studies[0],
study=study_graph_01).validate()
# Write the study as ISA-Tab txt file
with open(join(out_path, investigation.studies[0].info.path),
"wt",
newline="") as outputf:
StudyWriter.from_stream(study_or_assay=study_graph_01,
output_file=outputf).write()
# Create a corresponding Assay graph
# Create at least on samples, one output material and one collection process
# Unique names are required for unambiguous node identification
# Explicit header definition per node is currently required to enable export to ISA-Tab
sample_01 = models.Material(
type="Sample Name",
unique_name="S1-sample-0815-N1",
name="0815-N1",
extract_label=None,
characteristics=(),
comments=(),
factor_values=(),
material_type=None,
headers=[table_headers.SAMPLE_NAME],
)
data_file_01 = models.Material(
type="Raw Data File",
unique_name="S1-A1-0815-N1-DNA1-WES1_L???_???_R1.fastq.gz-COL4",
name="0815-N1-DNA1-WES1_L???_???_R1.fastq.gz",
extract_label=None,
characteristics=(),
comments=(),
factor_values=(),
material_type=None,
headers=[table_headers.RAW_DATA_FILE],
)
data_file_02 = models.Material(
type="Raw Data File",
unique_name="S1-A1-0815-N1-DNA1-WES1_L???_???_R2.fastq.gz-COL5",
name="0815-N1-DNA1-WES1_L???_???_R2.fastq.gz",
extract_label=None,
characteristics=(),
comments=(),
factor_values=(),
material_type=None,
headers=[table_headers.RAW_DATA_FILE],
)
process_01 = models.Process(
protocol_ref="nucleic acid sequencing",
unique_name="S1-A1-0815-N1-DNA1-WES1-3",
name="0815-N1-DNA1-WES1",
name_type="Assay Name",
date=None,
performer=None,
parameter_values=(),
comments=(),
array_design_ref=None,
first_dimension=None,
second_dimension=None,
headers=[table_headers.PROTOCOL_REF, table_headers.ASSAY_NAME],
)
# Create the arcs to connect the material and process nodes, referenced by the unique name
arcs = (
models.Arc(tail="S1-sample-0815-N1", head="S1-A1-0815-N1-DNA1-WES1-3"),
models.Arc(
tail="S1-A1-0815-N1-DNA1-WES1-3",
head="S1-A1-0815-N1-DNA1-WES1_L???_???_R1.fastq.gz-COL4",
),
models.Arc(
tail="S1-A1-0815-N1-DNA1-WES1_L???_???_R1.fastq.gz-COL4",
head="S1-A1-0815-N1-DNA1-WES1_L???_???_R2.fastq.gz-COL5",
),
)
# Create the assay graph object
assay_graph_01 = models.Assay(
file=investigation.studies[0].assays[0].path,
header=None,
materials={
sample_01.unique_name: sample_01,
data_file_01.unique_name: data_file_01,
data_file_02.unique_name: data_file_02,
},
processes={process_01.unique_name: process_01},
arcs=arcs,
)
# Validate assay graph
AssayValidator(
investigation=investigation,
study_info=investigation.studies[0],
assay_info=investigation.studies[0].assays[0],
assay=assay_graph_01,
).validate()
# Write the assay as ISA-Tab txt file
with open(join(out_path, investigation.studies[0].assays[0].path),
"wt",
newline="") as outputf:
AssayWriter.from_stream(study_or_assay=assay_graph_01,
output_file=outputf).write()