The immuneML ecosystem for machine learning analysis of adaptive immune receptor repertoires

The Observed T Cell Receptor Space database enables paired-chain repertoire mining, coherence analysis, and language modeling

T cell activation is governed through T cell receptors (TCRs), heterodimers of two sequence-variable chains (often an α and β chain) that synergistically recognize antigen fragments presented on cell surfaces. Despite this, there only exist repositories dedicated to collecting single-chain, not paired-chain, TCR sequence data. We addressed this gap by creating the Observed TCR Space (OTS) database, a source of consistently processed and annotated, full-length, paired-chain TCR sequences. Currently, OTS contains 5.35 million redundant (1.63 million non-redundant), predominantly human sequences from across 50 studies and at least 75 individuals. Using OTS, we identify pairing biases, public TCRs, and distinct chain coherence patterns relative to antibodies. We also release a paired-chain TCR language model, providing paired embedding representations and a method for residue in-filling conditional on the partner chain. OTS will be updated as a central community resource and is freely downloadable and available as a web application.

Biophysical cartography of the native and human-engineered antibody landscapes quantifies the plasticity of antibody developability

Designing effective monoclonal antibody (mAb) therapeutics faces a multi-parameter optimization challenge known as "developability", which reflects an antibody's ability to progress through development stages based on its physicochemical properties. While natural antibodies may provide valuable guidance for mAb selection, we lack a comprehensive understanding of natural developability parameter (DP) plasticity (redundancy, predictability, sensitivity) and how the DP landscapes of human-engineered and natural antibodies relate to one another. These gaps hinder fundamental developability profile cartography. To chart natural and engineered DP landscapes, we computed 40 sequence- and 46 structure-based DPs of over two million native and human-engineered single-chain antibody sequences. We find lower redundancy among structure-based compared to sequence-based DPs. Sequence DP sensitivity to single amino acid substitutions varied by antibody region and DP, and structure DP values varied across the conformational ensemble of antibody structures. We show that sequence DPs are more predictable than structure-based ones across different machine-learning tasks and embeddings, indicating a constrained sequence-based design space. Human-engineered antibodies localize within the developability and sequence landscapes of natural antibodies, suggesting that human-engineered antibodies explore mere subspaces of the natural one. Our work quantifies the plasticity of antibody developability, providing a fundamental resource for multi-parameter therapeutic mAb design.

Robust detection of infectious disease, autoimmunity, and cancer from the paratope networks of adaptive immune receptors

Liquid biopsies based on peripheral blood offer a minimally invasive alternative to solid tissue biopsies for the detection of diseases, primarily cancers. However, such tests currently consider only the serum component of blood, overlooking a potentially rich source of biomarkers: adaptive immune receptors (AIRs) expressed on circulating B and T cells. Machine learning-based classifiers trained on AIRs have been reported to accurately identify not only cancers but also autoimmune and infectious diseases as well. However, when using the conventional "clonotype cluster" representation of AIRs, individuals within a disease or healthy cohort exhibit vastly different features, limiting the generalizability of these classifiers. This study aimed to address the challenge of classifying specific diseases from circulating B or T cells by developing a novel representation of AIRs based on similarity networks constructed from their antigen-binding regions (paratopes). Features based on this novel representation, paratope cluster occupancies (PCOs), significantly improved disease classification performance for infectious disease, autoimmune disease, and cancer. Under identical methodological conditions, classifiers trained on PCOs achieved a mean AUC of 0.893 when applied to new individuals, outperforming clonotype cluster-based classifiers (AUC 0.714) and the best-performing published classifier (AUC 0.777). Surprisingly, for cancer patients, we observed that "healthy-biased" AIRs were predicted to target known cancer-associated antigens at dramatically higher rates than healthy AIRs as a whole (Z scores >75), suggesting an overlooked reservoir of cancer-targeting immune cells that could be identified by PCOs.

nf-core/airrflow: An adaptive immune receptor repertoire analysis workflow employing the Immcantation framework

Adaptive Immune Receptor Repertoire sequencing (AIRR-seq) is a valuable experimental tool to study the immune state in health and following immune challenges such as infectious diseases, (auto)immune diseases, and cancer. Several tools have been developed to reconstruct B cell and T cell receptor sequences from AIRR-seq data and infer B and T cell clonal relationships. However, currently available tools offer limited parallelization across samples, scalability or portability to high-performance computing infrastructures. To address this need, we developed nf-core/airrflow, an end-to-end bulk and single-cell AIRR-seq processing workflow which integrates the Immcantation Framework following BCR and TCR sequencing data analysis best practices. The Immcantation Framework is a comprehensive toolset, which allows the processing of bulk and single-cell AIRR-seq data from raw read processing to clonal inference. nf-core/airrflow is written in Nextflow and is part of the nf-core project, which collects community contributed and curated Nextflow workflows for a wide variety of analysis tasks. We assessed the performance of nf-core/airrflow on simulated sequencing data with sequencing errors and show example results with real datasets. To demonstrate the applicability of nf-core/airrflow to the high-throughput processing of large AIRR-seq datasets, we validated and extended previously reported findings of convergent antibody responses to SARS-CoV-2 by analyzing 97 COVID-19 infected individuals and 99 healthy controls, including a mixture of bulk and single-cell sequencing datasets. Using this dataset, we extended the convergence findings to 20 additional subjects, highlighting the applicability of nf-core/airrflow to validate findings in small in-house cohorts with reanalysis of large publicly available AIRR datasets.

Best practices for machine learning in antibody discovery and development

In the past 40 years, therapeutic antibody discovery and development have advanced considerably, with machine learning (ML) offering a promising way to speed up the process by reducing costs and the number of experiments required. Recent progress in ML-guided antibody design and development (D&D) has been hindered by the diversity of data sets and evaluation methods, which makes it difficult to conduct comparisons and assess utility. Establishing standards and guidelines will be crucial for the wider adoption of ML and the advancement of the field. This perspective critically reviews current practices, highlights common pitfalls and proposes method development and evaluation guidelines for various ML-based techniques in therapeutic antibody D&D. Addressing challenges across the ML process, best practices are recommended for each stage to enhance reproducibility and progress.

Enhancing comparative T cell receptor repertoire analysis in small biological samples through pooling homologous cell samples from multiple mice

Accurate characterization and comparison of T cell receptor (TCR) repertoires from small biological samples present significant challenges. The main challenge is the low material input, which compromises the quality of bulk sequencing and hinders the recovery of sufficient TCR sequences for robust analyses. We aimed to address this limitation by implementing a strategic approach to pool homologous biological samples. Our findings demonstrate that such pooling indeed enhances the TCR repertoire coverage, particularly for cell subsets of constrained sizes, and enables accurate comparisons of TCR repertoires at different levels of complexity across T cell subsets with different sizes. This methodology holds promise for advancing our understanding of T cell repertoires in scenarios where sample size constraints are a prevailing concern.

Guidelines for reproducible analysis of adaptive immune receptor repertoire sequencing data

Enhancing the reproducibility and comprehension of adaptive immune receptor repertoire sequencing (AIRR-seq) data analysis is critical for scientific progress. This study presents guidelines for reproducible AIRR-seq data analysis, and a collection of ready-to-use pipelines with comprehensive documentation. To this end, ten common pipelines were implemented using ViaFoundry, a user-friendly interface for pipeline management and automation. This is accompanied by versioned containers, documentation and archiving capabilities. The automation of pre-processing analysis steps and the ability to modify pipeline parameters according to specific research needs are emphasized. AIRR-seq data analysis is highly sensitive to varying parameters and setups; using the guidelines presented here, the ability to reproduce previously published results is demonstrated. This work promotes transparency, reproducibility, and collaboration in AIRR-seq data analysis, serving as a model for handling and documenting bioinformatics pipelines in other research domains.

Forum on immune digital twins: a meeting report

Medical digital twins are computational models of human biology relevant to a given medical condition, which are tailored to an individual patient, thereby predicting the course of disease and individualized treatments, an important goal of personalized medicine. The immune system, which has a central role in many diseases, is highly heterogeneous between individuals, and thus poses a major challenge for this technology. In February 2023, an international group of experts convened for two days to discuss these challenges related to immune digital twins. The group consisted of clinicians, immunologists, biologists, and mathematical modelers, representative of the interdisciplinary nature of medical digital twin development. A video recording of the entire event is available. This paper presents a synopsis of the discussions, brief descriptions of ongoing digital twin projects at different stages of progress. It also proposes a 5-year action plan for further developing this technology. The main recommendations are to identify and pursue a small number of promising use cases, to develop stimulation-specific assays of immune function in a clinical setting, and to develop a database of existing computational immune models, as well as advanced modeling technology and infrastructure.

Language model-based B cell receptor sequence embeddings can effectively encode receptor specificity

High throughput sequencing of B cell receptors (BCRs) is increasingly applied to study the immense diversity of antibodies. Learning biologically meaningful embeddings of BCR sequences is beneficial for predictive modeling. Several embedding methods have been developed for BCRs, but no direct performance benchmarking exists. Moreover, the impact of the input sequence length and paired-chain information on the prediction remains to be explored. We evaluated the performance of multiple embedding models to predict BCR sequence properties and receptor specificity. Despite the differences in model architectures, most embeddings effectively capture BCR sequence properties and specificity. BCR-specific embeddings slightly outperform general protein language models in predicting specificity. In addition, incorporating full-length heavy chains and paired light chain sequences improves the prediction performance of all embeddings. This study provides insights into the properties of BCR embeddings to improve downstream prediction applications for antibody analysis and discovery.

The dengue-specific immune response and antibody identification with machine learning

Dengue virus poses a serious threat to global health and there is no specific therapeutic for it. Broadly neutralizing antibodies recognizing all serotypes may be an effective treatment. High-throughput adaptive immune receptor repertoire sequencing (AIRR-seq) and bioinformatic analysis enable in-depth understanding of the B-cell immune response. Here, we investigate the dengue antibody response with these technologies and apply machine learning to identify rare and underrepresented broadly neutralizing antibody sequences. Dengue immunization elicited the following signatures on the antibody repertoire: (i) an increase of CDR3 and germline gene diversity; (ii) a change in the antibody repertoire architecture by eliciting power-law network distributions and CDR3 enrichment in polar amino acids; (iii) an increase in the expression of JNK/Fos transcription factors and ribosomal proteins. Furthermore, we demonstrate the applicability of computational methods and machine learning to AIRR-seq datasets for neutralizing antibody candidate sequence identification. Antibody expression and functional assays have validated the obtained results.

Assessing the feasibility of statistical inference using synthetic antibody-antigen datasets

Simulation frameworks are useful to stress-test predictive models when data is scarce, or to assert model sensitivity to specific data distributions. Such frameworks often need to recapitulate several layers of data complexity, including emergent properties that arise implicitly from the interaction between simulation components. Antibody-antigen binding is a complex mechanism by which an antibody sequence wraps itself around an antigen with high affinity. In this study, we use a synthetic simulation framework for antibody-antigen folding and binding on a 3D lattice that include full details on the spatial conformation of both molecules. We investigate how emergent properties arise in this framework, in particular the physical proximity of amino acids, their presence on the binding interface, or the binding status of a sequence, and relate that to the individual and pairwise contributions of amino acids in statistical models for binding prediction. We show that weights learnt from a simple logistic regression model align with some but not all features of amino acids involved in the binding, and that predictive sequence binding patterns can be enriched. In particular, main effects correlated with the capacity of a sequence to bind any antigen, while statistical interactions were related to sequence specificity.

An integrated approach to the characterization of immune repertoires using AIMS: An Automated Immune Molecule Separator

The adaptive immune system employs an array of receptors designed to respond with high specificity to pathogens or molecular aberrations faced by the host organism. Binding of these receptors to molecular fragments-collectively referred to as antigens-initiates immune responses. These antigenic targets are recognized in their native state on the surfaces of pathogens by antibodies, whereas T cell receptors (TCR) recognize processed antigens as short peptides, presented on major histocompatibility complex (MHC) molecules. Recent research has led to a wealth of immune repertoire data that are key to interrogating the nature of these molecular interactions. However, existing tools for the analysis of these large datasets typically focus on molecular sets of a single type, forcing researchers to separately analyze strongly coupled sequences of interacting molecules. Here, we introduce a software package for the integrated analysis of immune repertoire data, capable of identifying distinct biophysical differences in isolated TCR, MHC, peptide, antibody, and antigen sequence data. This integrated analytical approach allows for direct comparisons across immune repertoire subsets and provides a starting point for the identification of key interaction hotspots in complementary receptor-antigen pairs. The software (AIMS-Automated Immune Molecule Separator) is freely available as an open access package in GUI or command-line form.

Quantitative approaches for decoding the specificity of the human T cell repertoire

T cell receptor (TCR)-peptide-major histocompatibility complex (pMHC) interactions play a vital role in initiating immune responses against pathogens, and the specificity of TCRpMHC interactions is crucial for developing optimized therapeutic strategies. The advent of high-throughput immunological and structural evaluation of TCR and pMHC has provided an abundance of data for computational approaches that aim to predict favorable TCR-pMHC interactions. Current models are constructed using information on protein sequence, structures, or a combination of both, and utilize a variety of statistical learning-based approaches for identifying the rules governing specificity. This review examines the current theoretical, computational, and deep learning approaches for identifying TCR-pMHC recognition pairs, placing emphasis on each method's mathematical approach, predictive performance, and limitations.

Can we predict T cell specificity with digital biology and machine learning?

Recent advances in machine learning and experimental biology have offered breakthrough solutions to problems such as protein structure prediction that were long thought to be intractable. However, despite the pivotal role of the T cell receptor (TCR) in orchestrating cellular immunity in health and disease, computational reconstruction of a reliable map from a TCR to its cognate antigens remains a holy grail of systems immunology. Current data sets are limited to a negligible fraction of the universe of possible TCR-ligand pairs, and performance of state-of-the-art predictive models wanes when applied beyond these known binders. In this Perspective article, we make the case for renewed and coordinated interdisciplinary effort to tackle the problem of predicting TCR-antigen specificity. We set out the general requirements of predictive models of antigen binding, highlight critical challenges and discuss how recent advances in digital biology such as single-cell technology and machine learning may provide possible solutions. Finally, we describe how predicting TCR specificity might contribute to our understanding of the broader puzzle of antigen immunogenicity.

A novel approach to T-cell receptor beta chain (TCRB) repertoire encoding using lossless string compression

MOTIVATION

T-cell receptor beta chain (TCRB) repertoires are crucial for understanding immune responses. However, their high diversity and complexity present significant challenges in representation and analysis. The main motivation of this study is to develop a unified and compact representation of a TCRB repertoire that can efficiently capture its inherent complexity and diversity and allow for direct inference.

RESULTS

We introduce a novel approach to TCRB repertoire encoding and analysis, leveraging the Lempel-Ziv 76 algorithm. This approach allows us to create a graph-like model, identify-specific sequence features, and produce a new encoding approach for an individual's repertoire. The proposed representation enables various applications, including generation probability inference, informative feature vector derivation, sequence generation, a new measure for diversity estimation, and a new sequence centrality measure. The approach was applied to four large-scale public TCRB sequencing datasets, demonstrating its potential for a wide range of applications in big biological sequencing data.

AVAILABILITY AND IMPLEMENTATION

Python package for implementation is available https://github.com/MuteJester/LZGraphs.

Precision Vaccinology Approaches for the Development of Adjuvanted Vaccines Targeted to Distinct Vulnerable Populations

Infection persists as one of the leading global causes of morbidity and mortality, with particular burden at the extremes of age and in populations who are immunocompromised or suffer chronic co-morbid diseases. By focusing discovery and innovation efforts to better understand the phenotypic and mechanistic differences in the immune systems of diverse vulnerable populations, emerging research in precision vaccine discovery and development has explored how to optimize immunizations across the lifespan. Here, we focus on two key elements of precision vaccinology, as applied to epidemic/pandemic response and preparedness, including (a) selecting robust combinations of adjuvants and antigens, and (b) coupling these platforms with appropriate formulation systems. In this context, several considerations exist, including the intended goals of immunization (e.g., achieving immunogenicity versus lessening transmission), reducing the likelihood of adverse reactogenicity, and optimizing the route of administration. Each of these considerations is accompanied by several key challenges. On-going innovation in precision vaccinology will expand and target the arsenal of vaccine components for protection of vulnerable populations.

DagSim: Combining DAG-based model structure with unconstrained data types and relations for flexible, transparent, and modularized data simulation

Data simulation is fundamental for machine learning and causal inference, as it allows exploration of scenarios and assessment of methods in settings with full control of ground truth. Directed acyclic graphs (DAGs) are well established for encoding the dependence structure over a collection of variables in both inference and simulation settings. However, while modern machine learning is applied to data of an increasingly complex nature, DAG-based simulation frameworks are still confined to settings with relatively simple variable types and functional forms. We here present DagSim, a Python-based framework for DAG-based data simulation without any constraints on variable types or functional relations. A succinct YAML format for defining the simulation model structure promotes transparency, while separate user-provided functions for generating each variable based on its parents ensure simulation code modularization. We illustrate the capabilities of DagSim through use cases where metadata variables control shapes in an image and patterns in bio-sequences. DagSim is available as a Python package at PyPI. Source code and documentation are available at: https://github.com/uio-bmi/dagsim.

Understanding repertoire sequencing data through a multiscale computational model of the germinal center

Sequencing of B-cell and T-cell immune receptor repertoires helps us to understand the adaptive immune response, although it only provides information about the clonotypes (lineages) and their frequencies and not about, for example, their affinity or antigen (Ag) specificity. To further characterize the identified clones, usually with special attention to the particularly abundant ones (dominant), additional time-consuming or expensive experiments are generally required. Here, we present an extension of a multiscale model of the germinal center (GC) that we previously developed to gain more insight in B-cell repertoires. We compare the extent that these simulated repertoires deviate from experimental repertoires established from single GCs, blood, or tissue. Our simulations show that there is a limited correlation between clonal abundance and affinity and that there is large affinity variability among same-ancestor (same-clone) subclones. Our simulations suggest that low-abundance clones and subclones, might also be of interest since they may have high affinity for the Ag. We show that the fraction of plasma cells (PCs) with high B-cell receptor (BcR) mRNA content in the GC does not significantly affect the number of dominant clones derived from single GCs by sequencing BcR mRNAs. Results from these simulations guide data interpretation and the design of follow-up experiments.

The temporal behavior of the murine T cell receptor repertoire following Immunotherapy

Immunotherapy is now an essential tool for cancer treatment, and the unique features of an individual's T cell receptor repertoire are known to play a key role in its effectiveness. The repertoire, famously vast due to a cascade of cellular mechanisms, can be quantified using repertoire sequencing. In this study, we sampled the repertoire over several time points following treatment with anti-CTLA-4, in a syngeniec mouse model for colorectal cancer, generating a longitudinal dataset of T cell clones and their abundance. The dynamics of the repertoire can be observed in response to treatment over a period of four weeks, as clonal expansion of specific clones ascends and descends. The data made available here can be used to determine treatment and predict its effect, while also providing a unique look at the behavior of the immune system over time.

GENTLE: a novel bioinformatics tool for generating features and building classifiers from T cell repertoire cancer data

BACKGROUND

In the global effort to discover biomarkers for cancer prognosis, prediction tools have become essential resources. TCR (T cell receptor) repertoires contain important features that differentiate healthy controls from cancer patients or differentiate outcomes for patients being treated with different drugs. Considering, tools that can easily and quickly generate and identify important features out of TCR repertoire data and build accurate classifiers to predict future outcomes are essential.

RESULTS

This paper introduces GENTLE (GENerator of T cell receptor repertoire features for machine LEarning): an open-source, user-friendly web-application tool that allows TCR repertoire researchers to discover important features; to create classifier models and evaluate them with metrics; and to quickly generate visualizations for data interpretations. We performed a case study with repertoires of TRegs (regulatory T cells) and TConvs (conventional T cells) from healthy controls versus patients with breast cancer. We showed that diversity features were able to distinguish between the groups. Moreover, the classifiers built with these features could correctly classify samples ('Healthy' or 'Breast Cancer')from the TRegs repertoire when trained with the TConvs repertoire, and from the TConvs repertoire when trained with the TRegs repertoire.

CONCLUSION

The paper walks through installing and using GENTLE and presents a case study and results to demonstrate the application's utility. GENTLE is geared towards any researcher working with TCR repertoire data and aims to discover predictive features from these data and build accurate classifiers. GENTLE is available on https://github.com/dhiego22/gentle and https://share.streamlit.io/dhiego22/gentle/main/gentle.py .

Interpretable artificial intelligence model for accurate identification of medical conditions using immune repertoire

Underlying medical conditions, such as cancer, kidney disease and heart failure, are associated with a higher risk for severe COVID-19. Accurate classification of COVID-19 patients with underlying medical conditions is critical for personalized treatment decision and prognosis estimation. In this study, we propose an interpretable artificial intelligence model termed VDJMiner to mine the underlying medical conditions and predict the prognosis of COVID-19 patients according to their immune repertoires. In a cohort of more than 1400 COVID-19 patients, VDJMiner accurately identifies multiple underlying medical conditions, including cancers, chronic kidney disease, autoimmune disease, diabetes, congestive heart failure, coronary artery disease, asthma and chronic obstructive pulmonary disease, with an average area under the receiver operating characteristic curve (AUC) of 0.961. Meanwhile, in this same cohort, VDJMiner achieves an AUC of 0.922 in predicting severe COVID-19. Moreover, VDJMiner achieves an accuracy of 0.857 in predicting the response of COVID-19 patients to tocilizumab treatment on the leave-one-out test. Additionally, VDJMiner interpretively mines and scores V(D)J gene segments of the T-cell receptors that are associated with the disease. The identified associations between single-cell V(D)J gene segments and COVID-19 are highly consistent with previous studies. The source code of VDJMiner is publicly accessible at https://github.com/TencentAILabHealthcare/VDJMiner. The web server of VDJMiner is available at https://gene.ai.tencent.com/VDJMiner/.

Altered somatic hypermutation patterns in COVID-19 patients classifies disease severity

Introduction

The success of the human body in fighting SARS-CoV2 infection relies on lymphocytes and their antigen receptors. Identifying and characterizing clinically relevant receptors is of utmost importance.

Methods

We report here the application of a machine learning approach, utilizing B cell receptor repertoire sequencing data from severely and mildly infected individuals with SARS-CoV2 compared with uninfected controls.

Results

In contrast to previous studies, our approach successfully stratifies non-infected from infected individuals, as well as disease level of severity. The features that drive this classification are based on somatic hypermutation patterns, and point to alterations in the somatic hypermutation process in COVID-19 patients.

Discussion

These features may be used to build and adapt therapeutic strategies to COVID-19, in particular to quantitatively assess potential diagnostic and therapeutic antibodies. These results constitute a proof of concept for future epidemiological challenges.

simAIRR: simulation of adaptive immune repertoires with realistic receptor sequence sharing for benchmarking of immune state prediction methods

BACKGROUND

Machine learning (ML) has gained significant attention for classifying immune states in adaptive immune receptor repertoires (AIRRs) to support the advancement of immunodiagnostics and therapeutics. Simulated data are crucial for the rigorous benchmarking of AIRR-ML methods. Existing approaches to generating synthetic benchmarking datasets result in the generation of naive repertoires missing the key feature of many shared receptor sequences (selected for common antigens) found in antigen-experienced repertoires.

RESULTS

We demonstrate that a common approach to generating simulated AIRR benchmark datasets can introduce biases, which may be exploited for undesired shortcut learning by certain ML methods. To mitigate undesirable access to true signals in simulated AIRR datasets, we devised a simulation strategy (simAIRR) that constructs antigen-experienced-like repertoires with a realistic overlap of receptor sequences. simAIRR can be used for constructing AIRR-level benchmarks based on a range of assumptions (or experimental data sources) for what constitutes receptor-level immune signals. This includes the possibility of making or not making any prior assumptions regarding the similarity or commonality of immune state-associated sequences that will be used as true signals. We demonstrate the real-world realism of our proposed simulation approach by showing that basic ML strategies perform similarly on simAIRR-generated and real-world experimental AIRR datasets.

CONCLUSIONS

This study sheds light on the potential shortcut learning opportunities for ML methods that can arise with the state-of-the-art way of simulating AIRR datasets. simAIRR is available as a Python package: https://github.com/KanduriC/simAIRR.

Reference-based comparison of adaptive immune receptor repertoires

B and T cell receptor (immune) repertoires can represent an individual's immune history. While current repertoire analysis methods aim to discriminate between health and disease states, they are typically based on only a limited number of parameters. Here, we introduce immuneREF: a quantitative multidimensional measure of adaptive immune repertoire (and transcriptome) similarity that allows interpretation of immune repertoire variation by relying on both repertoire features and cross-referencing of simulated and experimental datasets. To quantify immune repertoire similarity landscapes across health and disease, we applied immuneREF to >2,400 datasets from individuals with varying immune states (healthy, [autoimmune] disease, and infection). We discovered, in contrast to the current paradigm, that blood-derived immune repertoires of healthy and diseased individuals are highly similar for certain immune states, suggesting that repertoire changes to immune perturbations are less pronounced than previously thought. In conclusion, immuneREF enables the population-wide study of adaptive immune response similarity across immune states.

Comparative Study of Repertoire Classification Methods Reveals Data Efficiency of k -mer Feature Extraction

The repertoire of T cell receptors encodes various types of immunological information. Machine learning is indispensable for decoding such information from repertoire datasets measured by next-generation sequencing (NGS). In particular, the classification of repertoires is the most basic task, which is relevant for a variety of scientific and clinical problems. Supported by the recent appearance of large datasets, efficient but data-expensive methods have been proposed. However, it is unclear whether they can work efficiently when the available sample size is severely restricted as in practical situations. In this study, we demonstrate that their performances can be impaired substantially below critical sample sizes. To complement this drawback, we propose MotifBoost, which exploits the information of short k-mer motifs of TCRs. MotifBoost can perform the classification as efficiently as a deep learning method on large datasets while providing more stable and reliable results on small datasets. We tested MotifBoost on the four small datasets which consist of various conditions such as Cytomegalovirus (CMV), HIV, α-chain, β-chain and it consistently preserved the stability. We also clarify that the robustness of MotifBoost can be attributed to the efficiency of k-mer motifs as representation features of repertoires. Finally, by comparing the predictions of these methods, we show that the whole sequence identity and sequence motifs encode partially different information and that a combination of such complementary information is necessary for further development of repertoire analysis.

CompAIRR: ultra-fast comparison of adaptive immune receptor repertoires by exact and approximate sequence matching

MOTIVATION

Adaptive immune receptor (AIR) repertoires (AIRRs) record past immune encounters with exquisite specificity. Therefore, identifying identical or similar AIR sequences across individuals is a key step in AIRR analysis for revealing convergent immune response patterns that may be exploited for diagnostics and therapy. Existing methods for quantifying AIRR overlap scale poorly with increasing dataset numbers and sizes. To address this limitation, we developed CompAIRR, which enables ultra-fast computation of AIRR overlap, based on either exact or approximate sequence matching.

RESULTS

CompAIRR improves computational speed 1000-fold relative to the state of the art and uses only one-third of the memory: on the same machine, the exact pairwise AIRR overlap of 104 AIRRs with 105 sequences is found in ∼17 min, while the fastest alternative tool requires 10 days. CompAIRR has been integrated with the machine learning ecosystem immuneML to speed up commonly used AIRR-based machine learning applications.

AVAILABILITY AND IMPLEMENTATION

CompAIRR code and documentation are available at https://github.com/uio-bmi/compairr. Docker images are available at https://hub.docker.com/r/torognes/compairr. The code to replicate the synthetic datasets, scripts for benchmarking and creating figures, and all raw data underlying the figures are available at https://github.com/uio-bmi/compairr-benchmarking.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Machine-designed biotherapeutics: opportunities, feasibility and advantages of deep learning in computational antibody discovery

Antibodies are versatile molecular binders with an established and growing role as therapeutics. Computational approaches to developing and designing these molecules are being increasingly used to complement traditional lab-based processes. Nowadays, in silico methods fill multiple elements of the discovery stage, such as characterizing antibody-antigen interactions and identifying developability liabilities. Recently, computational methods tackling such problems have begun to follow machine learning paradigms, in many cases deep learning specifically. This paradigm shift offers improvements in established areas such as structure or binding prediction and opens up new possibilities such as language-based modeling of antibody repertoires or machine-learning-based generation of novel sequences. In this review, we critically examine the recent developments in (deep) machine learning approaches to therapeutic antibody design with implications for fully computational antibody design.

Machine Learning Approaches to TCR Repertoire Analysis

Sparked by the development of genome sequencing technology, the quantity and quality of data handled in immunological research have been changing dramatically. Various data and database platforms are now driving the rapid progress of machine learning for immunological data analysis. Of various topics in immunology, T cell receptor repertoire analysis is one of the most important targets of machine learning for assessing the state and abnormalities of immune systems. In this paper, we review recent repertoire analysis methods based on machine learning and deep learning and discuss their prospects.

Profiling the baseline performance and limits of machine learning models for adaptive immune receptor repertoire classification

BACKGROUND

Machine learning (ML) methodology development for the classification of immune states in adaptive immune receptor repertoires (AIRRs) has seen a recent surge of interest. However, so far, there does not exist a systematic evaluation of scenarios where classical ML methods (such as penalized logistic regression) already perform adequately for AIRR classification. This hinders investigative reorientation to those scenarios where method development of more sophisticated ML approaches may be required.

RESULTS

To identify those scenarios where a baseline ML method is able to perform well for AIRR classification, we generated a collection of synthetic AIRR benchmark data sets encompassing a wide range of data set architecture-associated and immune state-associated sequence patterns (signal) complexity. We trained ≈1,700 ML models with varying assumptions regarding immune signal on ≈1,000 data sets with a total of ≈250,000 AIRRs containing ≈46 billion TCRβ CDR3 amino acid sequences, thereby surpassing the sample sizes of current state-of-the-art AIRR-ML setups by two orders of magnitude. We found that L1-penalized logistic regression achieved high prediction accuracy even when the immune signal occurs only in 1 out of 50,000 AIR sequences.

CONCLUSIONS

We provide a reference benchmark to guide new AIRR-ML classification methodology by (i) identifying those scenarios characterized by immune signal and data set complexity, where baseline methods already achieve high prediction accuracy, and (ii) facilitating realistic expectations of the performance of AIRR-ML models given training data set properties and assumptions. Our study serves as a template for defining specialized AIRR benchmark data sets for comprehensive benchmarking of AIRR-ML methods.

TCRpower: quantifying the detection power of T-cell receptor sequencing with a novel computational pipeline calibrated by spike-in sequences

T-cell receptor (TCR) sequencing has enabled the development of innovative diagnostic tests for cancers, autoimmune diseases and other applications. However, the rarity of many T-cell clonotypes presents a detection challenge, which may lead to misdiagnosis if diagnostically relevant TCRs remain undetected. To address this issue, we developed TCRpower, a novel computational pipeline for quantifying the statistical detection power of TCR sequencing methods. TCRpower calculates the probability of detecting a TCR sequence as a function of several key parameters: in-vivo TCR frequency, T-cell sample count, read sequencing depth and read cutoff. To calibrate TCRpower, we selected unique TCRs of 45 T-cell clones (TCCs) as spike-in TCRs. We sequenced the spike-in TCRs from TCCs, together with TCRs from peripheral blood, using a 5' RACE protocol. The 45 spike-in TCRs covered a wide range of sample frequencies, ranging from 5 per 100 to 1 per 1 million. The resulting spike-in TCR read counts and ground truth frequencies allowed us to calibrate TCRpower. In our TCR sequencing data, we observed a consistent linear relationship between sample and sequencing read frequencies. We were also able to reliably detect spike-in TCRs with frequencies as low as one per million. By implementing an optimized read cutoff, we eliminated most of the falsely detected sequences in our data (TCR α-chain 99.0% and TCR β-chain 92.4%), thereby improving diagnostic specificity. TCRpower is publicly available and can be used to optimize future TCR sequencing experiments, and thereby enable reliable detection of disease-relevant TCRs for diagnostic applications.

In silico proof of principle of machine learning-based antibody design at unconstrained scale

Generative machine learning (ML) has been postulated to become a major driver in the computational design of antigen-specific monoclonal antibodies (mAb). However, efforts to confirm this hypothesis have been hindered by the infeasibility of testing arbitrarily large numbers of antibody sequences for their most critical design parameters: paratope, epitope, affinity, and developability. To address this challenge, we leveraged a lattice-based antibody-antigen binding simulation framework, which incorporates a wide range of physiological antibody-binding parameters. The simulation framework enables the computation of synthetic antibody-antigen 3D-structures, and it functions as an oracle for unrestricted prospective evaluation and benchmarking of antibody design parameters of ML-generated antibody sequences. We found that a deep generative model, trained exclusively on antibody sequence (one dimensional: 1D) data can be used to design conformational (three dimensional: 3D) epitope-specific antibodies, matching, or exceeding the training dataset in affinity and developability parameter value variety. Furthermore, we established a lower threshold of sequence diversity necessary for high-accuracy generative antibody ML and demonstrated that this lower threshold also holds on experimental real-world data. Finally, we show that transfer learning enables the generation of high-affinity antibody sequences from low-N training data. Our work establishes a priori feasibility and the theoretical foundation of high-throughput ML-based mAb design.

Individualized VDJ recombination predisposes the available Ig sequence space

The process of recombination between variable (V), diversity (D), and joining (J) immunoglobulin (Ig) gene segments determines an individual's naive Ig repertoire and, consequently, (auto)antigen recognition. VDJ recombination follows probabilistic rules that can be modeled statistically. So far, it remains unknown whether VDJ recombination rules differ between individuals. If these rules differed, identical (auto)antigen-specific Ig sequences would be generated with individual-specific probabilities, signifying that the available Ig sequence space is individual specific. We devised a sensitivity-tested distance measure that enables inter-individual comparison of VDJ recombination models. We discovered, accounting for several sources of noise as well as allelic variation in Ig sequencing data, that not only unrelated individuals but also human monozygotic twins and even inbred mice possess statistically distinguishable immunoglobulin recombination models. This suggests that, in addition to genetic, there is also nongenetic modulation of VDJ recombination. We demonstrate that population-wide individualized VDJ recombination can result in orders of magnitude of difference in the probability to generate (auto)antigen-specific Ig sequences. Our findings have implications for immune receptor-based individualized medicine approaches relevant to vaccination, infection, and autoimmunity.