Paper detail

New algorithmic challenges of adaptive immune repertoire construction

Motivation: The analysis of antibodies and T-cell receptors (TCRs) concentrations in serum is a fundamental problem in immunoinformatics. Repertoire construction is a preliminary step of analysis of clonal lineages, understanding of immune response dynamics, population analysis of immunoglobulin and TCR loci. Emergence of MiSeq Illumina sequencing machine in 2013 opened horizons of investigation of adaptive immune repertoires using highly accurate reads. Reads produced by MiSeq are able to cover repertoires of moderate size. At the same time, throughput of sequencing machines increases from year to year. This will enable ultra deep scanning of adaptive immune repertoires and analysis of their diversity. Such data requires both efficient and highly accurate repertoire construction tools. In 2015 Safonova et al. presented IgRepertoireConstructor, a tool for accurate construction of antibody repertoire and immunoproteogenomics analysis. Unfortunately, proposed algorithm was very time and memory consuming and could be a bottleneck of processing large immunosequencing libraries. In this paper we overcome this challenge and present IgReC, a novel algorithm for adaptive repertoire construction problem. IgReC reconstructs a repertoire with high precision even if each input read contains sequencing errors and performs well on contemporary datasets. Results of computational experiments show that IgReC improves state-of-the-art in the field. Availability: IgReC is an open source and freely available program running on Linux platforms. The source code is available at GitHub: yana-safonova.github.io/ig_repertoire_constructor. Contact: safonova.yana@gmail.com