Novel Methods for Correcting Next Generation Sequencing Errors in the β Chain of T Cell Receptors

摘要

Next generation sequencing (NGS) promises to revolutionize the various areas of the omics family by skyrocketing the throughput by which sequencing data become available. In immunogenetics, substantial effort has been put on the analysis of the repertoire of T cell receptors (TCR), one of the main weapons of the adaptive immune system, since this analysis is of great significance for the study of lymphoid malignancies, such as the Chronic Lymphocytic Leukemia (CLL). However, sequencing errors, inherent even in traditional low-throughput sequencing, can interfere with repertoire analysis, possibly misleading the immunogenetic interpretation of the analysis outcomes. The high sequencing error rate of NGS aggravates the problem even further. In this paper, we propose two novel error correction methods that are applied sequentially to NGS data acquired from the TCR β chain. Both methods focus on the highly variable Complementarity Determining Region 3 (CDR3) of the TCR. First, we attempt to revise the boundaries of the non-template parts of CDR3 that have been extracted by a third-party immunogenetic annotation tool, and then we correct the V- and J-gene parts of the CDR3 based on template germline V- and J-genes. The proposed methods have undergone preliminary evaluation on TCR β chain NGS data of two CLL patients, yielding promising results.