Decombinator: a tool for fast, efficient gene assignment in T-cell receptor sequences using a finite state machine

Area of Science:

• Immunology
• Bioinformatics
• Genomics

Background:

• High-throughput sequencing generates vast amounts of data for analyzing antigen-specific receptors like T-cell receptors (TcRs).
• Efficient data categorization and storage are crucial for subsequent analysis of this complex repertoire.

Purpose of the Study:

• To develop a novel method for uniquely identifying and categorizing individual TcR sequences from high-throughput sequencing data.
• To create a software package, Decombinator, that efficiently maps short-read sequence data to these identifiers, accommodating sequencing errors.

Main Methods:

• Defined a simple five-item identifier for unambiguous TcR sequence definition.
• Applied a novel finite-state automaton approach to map Illumina short-read data to TcR identifiers.
• Extended the algorithm to handle single-base pair mismatches common in sequencing data.

Main Results:

• Decombinator achieved analysis rates over two orders of magnitude faster than classical pairwise alignment algorithms.
• Demonstrated high accuracy (>88%) in assigning TcR sequences, even with up to 1% introduced error rates.
• Analysis revealed significant V and J gene usage bias in the human peripheral blood T-cell repertoire, independent of antigen exposure.

Conclusions:

• The Decombinator package offers a valuable tool for rapid and accurate analysis of the T-cell receptor repertoire.
• Efficiently handling large-scale sequencing data is essential for understanding the vastness and complexity of immune system responses.
• Further in-depth analysis of the T-cell repertoire is facilitated by tools like Decombinator.