Computational Strategies for Dissecting the High-Dimensional Complexity of Adaptive Immune Repertoires

Abstract

The adaptive immune system recognizes antigens via an immense array of antigen-binding antibodies and T-cell receptors, the immune repertoire. The interrogation of immune repertoires is of high relevance for understanding the adaptive immune response in disease and infection (e.g., autoimmunity, cancer, HIV). Adaptive immune receptor repertoire sequencing (AIRR-seq) has driven the quantitative and molecular-level profiling of immune repertoires, thereby revealing the high-dimensional complexity of the immune receptor sequence landscape. Several methods for the computational and statistical analysis of large-scale AIRR-seq data have been developed to resolve immune repertoire complexity and to understand the dynamics of adaptive immunity. Here, we review the current research on (i) diversity, (ii) clustering and network, (iii) phylogenetic, and (iv) machine learning methods applied to dissect, quantify, and compare the architecture, evolution, and specificity of immune repertoires. We summarize outstanding questions in computational immunology and propose future directions for systems immunology toward coupling AIRR-seq with the computational discovery of immunotherapeutics, vaccines, and immunodiagnostics.

Keywords: B-cell receptor; T-cell receptor; antibody discovery; artificial intelligence; immunogenomics; networks; phylogenetics; systems immunology.

Figure 1

The immune repertoire space is defined by diversity, architecture, evolution, and convergence. (A) Diversity measurements are based on (i) the accurate annotation of V (D) J segments using deterministic and probabilistic approaches with population-level or individualized germline gene reference databases. (ii) Probabilistic and hidden Markov models allow inference of recombination statistics. (iii) Measurement of clonotype diversity using diversity profiles. (B) Analysis of repertoire architecture relies predominantly on (i) clonal networks that are constructed by connecting nucleotide or amino acid sequence nodes by similarity edges. The sequence similarity between clones is defined via a string distance [e.g., Levenshtein distance (LD)], resulting in undirected Boolean networks for a given threshold (nucleotides/amino acids). An example of the global characterization of the network is the diameter, shown by black edges. An example of the local parameters of the network is the degree (n = 1) related to the individual clonal node in black. (ii) Degree distribution is a global characteristic of immune repertoire networks, which can be used for analyzing clonal expansion. (iii) Several similarity layers decompose the immune repertoire along its similarity layers. Layer D1 captures clonal nodes similar by edit distance 1 (1 nt/a.a. different), D2 of distance 2 and so forth. (C) Assessing evolution of antibody lineages. (i) Reconstruction of phylogenetic trees. Stars indicate somatic hypermutation. (ii) Probabilistic methods for the inference of mutation statistics in antibody lineage evolution. (iii) Simulation of antibody repertoire evolution for benchmarking antibody-tailored phylogenetic inference algorithms. (D) Naive and antigen-driven cross-individual sequence similarity and convergence in immune repertoires. (i) The Venn diagram shows sequences shared in the two repertoires (circles). Signature-like sequence features are highlighted by black squares. (ii) Database of convergent or antigen-specific immune receptor sequences. (iii) K-mer sequence decomposition and classification of immune receptor sequences.

Figure 2

An overview of selected computational tools used in immune repertoire analyses. Each horizontal colored bar colored bar in the Basis column represents a unique antibody or T-cell receptor (TCR) sequence. Vertical red bars represent sequence differences or somatic hypermutation. The Method column describes the general concept of the computational methods and how these are applied to immune repertoires. The Tools column highlights exemplary key resources for performing computational analysis in the respective analytical sections [rows (A–D)].