Re-synchronize biomarker evidence items/classifications with a (possibly further-filtered) variant set
Source:R/variant_classification.R
sync_biomarker_evidence.RdBiomarker evidence matching (map_biomarker_data(),
assign_amp_asco_cap_tiers()) is performed once, early, against
the variant set as it existed right after annotation. Callers may go on
to remove variants from that set afterwards (e.g. allelic depth/fraction
filtering via filter_read_support(), or germline/non-exonic
filtering for tumor-only input). Without re-syncing, evidence items for
variants removed by such downstream filtering remain "orphaned" -
present in bm_evidence but absent from the variant set - which
surfaces as biomarker-matched variants missing from the variant listing
in the report while still appearing in the biomarker evidence listing
(see https://github.com/sigven/pcgr/issues/302).
Arguments
- bm_evidence
list with biomarker evidence data (as initialized by
init_biomarker_content()), i.e. top-level 'eitems'/'classification' data frames plus one sub-list per clinical significance category, each with its own 'eitems'/'classification' data frames- var_df
data frame with the final (filtered) variant set