Identify clonotypes at multiple stringency levels

Groups cells into clonotypes using a hierarchy of stringency levels, from strict (all four chains must match) to permissive (a single allele suffices). Uses graph-based clustering via igraph to find connected components of matching cells. Supports parallel processing across samples.

Usage

identify_clonotypes(
  TCR_data,
  clone_definition_df = .clone.definition.df,
  num_cores = 1,
  sample_col = NULL,
  clone_loose = "single_chain_single_allele",
  verbose = TRUE
)

Arguments

TCR_data

A data frame with columns TRA_1, TRA_2, TRB_1, TRB_2 containing validated TCR chain identifiers.

clone_definition_df

Clone definition data frame specifying the stringency hierarchy. Defaults to the built-in definition.

num_cores

Number of cores for parallel processing across samples. Set to -1 to use all available cores. The actual number of cores used is capped at the number of available cores and the number of unique samples.

sample_col

Name of the column to use as sample identifier. If NULL, all cells are treated as one sample.

clone_loose

Which stringency level to use as the "loose" default. One of: "dual_chain_dual_allele", "dual_chain_one_partial", "dual_chain_both_partial", "single_chain_dual_allele", "single_chain_single_allele".

verbose

Logical. If TRUE, print progress messages.

Value

A list with two elements:

TCR_data

Data frame with clone ID columns for each stringency level (CloneID.dual_chain_dual_allele, etc.) plus CloneID.loose.

ref_tables

Named list of reference tables for each stringency level, each containing CloneID, Sample, and unique TCR chain info.