def get_gene_to_haplotypes_call(
cls, full_call_data: FullCallData,
panel: Panel) -> Dict[str, Set[HaplotypeCall]]:
gene_to_haplotype_calls = {}
for gene_info in panel.get_gene_infos():
logging.info(f"Calling haplotypes for {gene_info.gene}")
gene_to_haplotype_calls[
gene_info.gene] = cls.__get_haplotypes_call(
full_call_data, gene_info)
return gene_to_haplotype_calls
def get_genotype_tsv_text(cls, pgx_analysis: PgxAnalysis, panel: Panel,
version: str) -> str:
gene_to_haplotype_calls = pgx_analysis.get_gene_to_haplotype_calls()
genes_in_analysis = set(gene_to_haplotype_calls.keys())
assert genes_in_analysis == panel.get_genes(), (
f"Gene lists inconsistent.\n"
f"From analysis={sorted(list(genes_in_analysis))}\n"
f"From panel={sorted(list(panel.get_genes()))}")
gene_to_drug_info = {}
for gene_info in panel.get_gene_infos():
sorted_drugs = sorted([drug for drug in gene_info.drugs],
key=lambda info:
(info.name, info.url_prescription_info))
gene_to_drug_info[gene_info.gene] = (cls.DRUG_SEPARATOR.join(
[drug.name for drug in sorted_drugs]),
cls.DRUG_SEPARATOR.join([
drug.url_prescription_info
for drug in sorted_drugs
]))
header = cls.TSV_SEPARATOR.join(cls.GENOTYPE_TSV_COLUMNS)
lines = [header]
for gene in sorted(gene_to_haplotype_calls.keys()):
if gene_to_haplotype_calls[gene]:
for haplotype_call in sorted(
gene_to_haplotype_calls[gene],
key=lambda call: call.haplotype_name):
lines.append(
cls.TSV_SEPARATOR.join([
gene,
haplotype_call.haplotype_name,
cls.__get_zygosity(haplotype_call),
panel.get_haplotype_function(
gene, haplotype_call.haplotype_name),
gene_to_drug_info[gene][0],
gene_to_drug_info[gene][1],
panel.get_id(),
version,
]))
else:
lines.append(
cls.TSV_SEPARATOR.join([
gene,
cls.UNRESOLVED_HAPLOTYPE_STRING,
cls.NOT_APPLICABLE_ZYGOSITY_STRING,
UNKNOWN_FUNCTION_STRING,
gene_to_drug_info[gene][0],
gene_to_drug_info[gene][1],
panel.get_id(),
version,
]))
text = "\n".join(lines) + "\n"
return text
def __get_calls_for_panel_variants_without_calls(
cls, simple_call_data: SimpleCallData,
panel: Panel) -> FrozenSet[SimpleCall]:
# assume ref call when no call is found. Set filter to NO_CALL
reference_assembly = simple_call_data.reference_assembly
rs_ids_found_in_patient = {
rs_id
for call in simple_call_data.calls for rs_id in call.rs_ids
if rs_id != "."
}
coordinates_covered_by_found_calls = {
coordinate
for call in simple_call_data.calls
for coordinate in call.get_relevant_coordinates()
}
uncalled_calls = set()
for gene_info in panel.get_gene_infos():
for rs_id_info in gene_info.rs_id_infos:
coordinates_partially_handled = bool(
rs_id_info.get_relevant_coordinates(reference_assembly).
intersection(coordinates_covered_by_found_calls))
if rs_id_info.rs_id not in rs_ids_found_in_patient and not coordinates_partially_handled:
# Assuming REF/REF relative to reference assembly
start_coordinate = rs_id_info.get_start_coordinate(
reference_assembly)
reference_allele = rs_id_info.get_reference_allele(
reference_assembly)
uncalled_ref_call = SimpleCall(
start_coordinate,
reference_allele,
(reference_allele, reference_allele),
gene_info.gene,
(rs_id_info.rs_id, ),
REF_CALL_ANNOTATION_STRING,
SimpleCallFilter.NO_CALL,
)
uncalled_calls.add(uncalled_ref_call)
return frozenset(uncalled_calls)