Compute similarity scores for all disagree rows in resolution_state

For each disagree row, scores each candidate DTXSID by comparing the user's original Name-tagged column value against the candidate's preferredName and all cached synonyms. Returns the best (max) candidate score per row. Non-disagree rows get NA_real_.

Usage

compute_similarity_scores(
  resolution_state,
  enrichment_cache,
  dtxsid_cols,
  column_tags
)

Arguments

resolution_state

Data frame with consensus_status, dtxsid_, preferredName_, rank_* columns

enrichment_cache

Tibble with dtxsid and (optionally) synonyms columns

dtxsid_cols

Character vector of dtxsid column names (e.g., c("dtxsid_Chemical", "dtxsid_CAS"))

column_tags

Named list (col_name -> "Name"|"CASRN"|"Other") – used to find input name column

Value

resolution_state with similarity_score column added