Find significant TCRdist matches between query and database TCRs

Computes paired TCRdist distances between query and database TCRs and converts them to p-values adjusted for both the number of query and database TCRs. Background distributions are estimated via the V(D)J rearrangement model in .estimate_background_tcrdist_distributions().

Usage

find_significant_tcrdist_matches(
  query_tcrs_df,
  db_tcrs_df,
  organism,
  adjusted_pvalue_threshold = 1,
  background_tcrs_df = NULL,
  num_random_samples = 50000L,
  fixup_alleles = TRUE
)

Arguments

query_tcrs_df

A data.frame with at least columns va (or va_gene), cdr3a, vb (or vb_gene), cdr3b. If background_tcrs_df is NULL, also needs ja, jb, cdr3a_nucseq, cdr3b_nucseq.

db_tcrs_df

A data.frame with at least columns va (or va_gene), cdr3a, vb (or vb_gene), cdr3b.

organism

Character string (e.g. "human", "mouse").

adjusted_pvalue_threshold

Numeric. Maximum adjusted p-value to report. Default 1.0.

background_tcrs_df

Optional data.frame for background generation. If NULL, uses query_tcrs_df (which must then include ja, jb, and nucseq columns).

num_random_samples

Integer. Number of random background samples. Default 50000L.

fixup_alleles

Logical. If TRUE, optimize allele assignments in background TCRs. Default TRUE.

Value

A data.frame with columns tcrdist, pvalue_adj, fdr_value, query_index (0-based), db_index (0-based), plus query and db TCR information. Sorted by pvalue_adj.

See also

match_tcrs_to_db, strict_single_chain_match_tcrs_to_db

Examples

if (FALSE) { # \dontrun{
matches <- find_significant_tcrdist_matches(
    query_tcrs_df = query_df,
    db_tcrs_df = db_df,
    organism = "human",
    adjusted_pvalue_threshold = 0.05
)
} # }