tcrdist3

tcrdist3 computes distance-based comparisons of T cell receptor (TCR) sequences in Python to quantify repertoire similarity and identify antigen-associated receptors from paired- and single-chain TCR sequencing data.

Key Features:

• Sequence Similarity Networks: Constructs and integrates sequence similarity networks to visualize relationships among TCR sequences and to identify shared motifs and critical contact residues among TCRs targeting the same epitope.
• Gene-Usage Plots and Background-Adjusted CDR3 Logos: Produces gene-usage plots and background-adjusted CDR3 sequence logos to highlight sequence features associated with antigen specificity, including analyses of peptide-MHC-multimer sorted receptors.
• TCRjoin: Implements a flexible querying mechanism to match receptor sequences of interest against bulk repertoires or libraries of previously annotated TCRs.
• TCRdist Metric and Neighborhood Analysis: Implements the TCRdist metric for distance calculation and supports identification of candidate polyclonal receptors via sequence neighbor enrichment testing with flexible neighborhood definitions (analogous to TCRNET and ALICE).

Scientific Applications:

• Antigen specificity mapping: Identification of sequence motifs and contact residues that underlie TCR recognition of specific epitopes.
• Detection of antigen-driven selection: Discovery of candidate polyclonal receptors under antigenic selection in bulk repertoires using neighbor enrichment analyses.
• Repertoire annotation and comparison: Comparison of receptors against annotated TCR libraries and bulk repertoires to annotate putative specificities.
• Repertoire diversity and dynamics analysis: Quantification and comparison of TCR repertoire similarity and diversity from paired- and single-chain TCR sequencing datasets.

Methodology:

Performs distance-based analysis using the TCRdist metric; constructs sequence similarity networks; generates gene-usage plots and background-adjusted CDR3 logos; matches sequences via TCRjoin; and applies sequence neighbor enrichment testing comparable to TCRNET and ALICE with user-definable neighborhood definitions.