Build display data for potential two-hit events (nested reactable)

For CNA records flagged with somatic and/or germline loss-of-function SNV/InDel candidates, this function parses the comma-separated TWOHIT_CANDIDATE_SOMATIC / TWOHIT_CANDIDATE_GERMLINE columns.

Usage

build_twohit_display_data(
  cna_variant = NULL,
  snv_somatic = NULL,
  snv_germline = NULL,
  settings = NULL
)

Arguments

cna_variant: data frame - gene-level CNA callset (pcg_report$content$cna$callset$variant)
snv_somatic: data frame - somatic SNV/InDel callset (retained for backward compatibility; no longer used for the somatic join)
snv_germline: data frame - germline classified callset (pcg_report$content$germline_classified$callset$variant)
settings: PCGR settings list (pcg_report$settings); used to apply tumor_dp_min and tumor_af_min thresholds

Value

Named list with elements main and nested (both data frames, empty if no two-hit candidates found).

Details

Somatic entry format (semicolon-separated): VAR_ID;CONSEQUENCE;VAF_FLAG;ALTERATION;VAF_TUMOR_PCT;ONCOGENICITY

Display fields for somatic variants are embedded by the Python pipeline from the unfiltered somatic callset, so variants below the depth filter still render correctly without a secondary join.

main: One row per two-hit CNA gene: SYMBOL, GENENAME, VARIANT_CLASS, CN_TOTAL, LOH, and a hidden .row_id key.
nested: One row per overlapping LoF variant: .row_id (FK), ORIGIN (Somatic / Germline), ALTERATION, CONSEQUENCE, VAF_GENOTYPE, VAF_FLAG, CLASSIFICATION.