DeepTCR is a deep learning framework for revealing sequence concepts within T-cell repertoires

Artificial intelligence in cancer immunotherapy: Applications in neoantigen recognition, antibody design and immunotherapy response prediction

Cancer immunotherapy is a method of controlling and eliminating tumors by reactivating the body's cancer-immunity cycle and restoring its antitumor immune response. The increased availability of data, combined with advancements in high-performance computing and innovative artificial intelligence (AI) technology, has resulted in a rise in the use of AI in oncology research. State-of-the-art AI models for functional classification and prediction in immunotherapy research are increasingly used to support laboratory-based experiments. This review offers a glimpse of the current AI applications in immunotherapy, including neoantigen recognition, antibody design, and prediction of immunotherapy response. Advancing in this direction will result in more robust predictive models for developing better targets, drugs, and treatments, and these advancements will eventually make their way into the clinical setting, pushing AI forward in the field of precision oncology.

Inferring the T cell repertoire dynamics of healthy individuals

The adaptive immune system is a diverse ecosystem that responds to pathogens by selecting cells with specific receptors. While clonal expansion in response to particular immune challenges has been extensively studied, we do not know the neutral dynamics that drive the immune system in the absence of strong stimuli. Here, we learn the parameters that underlie the clonal dynamics of the T cell repertoire in healthy individuals of different ages, by applying Bayesian inference to longitudinal immune repertoire sequencing (RepSeq) data. Quantifying the experimental noise accurately for a given RepSeq technique allows us to disentangle real changes in clonal frequencies from noise. We find that the data are consistent with clone sizes following a geometric Brownian motion and show that its predicted steady state is in quantitative agreement with the observed power-law behavior of the clone-size distribution. The inferred turnover time scale of the repertoire increases with patient age and depends on the clone size in some individuals.

Counting is almost all you need

The immune memory repertoire encodes the history of present and past infections and immunological attributes of the individual. As such, multiple methods were proposed to use T-cell receptor (TCR) repertoires to detect disease history. We here show that the counting method outperforms two leading algorithms. We then show that the counting can be further improved using a novel attention model to weigh the different TCRs. The attention model is based on the projection of TCRs using a Variational AutoEncoder (VAE). Both counting and attention algorithms predict better than current leading algorithms whether the host had CMV and its HLA alleles. As an intermediate solution between the complex attention model and the very simple counting model, we propose a new Graph Convolutional Network approach that obtains the accuracy of the attention model and the simplicity of the counting model. The code for the models used in the paper is provided at: https://github.com/louzounlab/CountingIsAlmostAllYouNeed.

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/.

Finding antigens for TB vaccines: the good, the bad and the useless

Prospective, longitudinal clinical studies incorporating high-throughput, single-cell analyses could identify which bacterial antigens to include in TB vaccines — and which to avoid.

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.

Machine learning methods for protein-protein binding affinity prediction in protein design

Protein-protein interactions govern a wide range of biological activity. A proper estimation of the protein-protein binding affinity is vital to design proteins with high specificity and binding affinity toward a target protein, which has a variety of applications including antibody design in immunotherapy, enzyme engineering for reaction optimization, and construction of biosensors. However, experimental and theoretical modelling methods are time-consuming, hinder the exploration of the entire protein space, and deter the identification of optimal proteins that meet the requirements of practical applications. In recent years, the rapid development in machine learning methods for protein-protein binding affinity prediction has revealed the potential of a paradigm shift in protein design. Here, we review the prediction methods and associated datasets and discuss the requirements and construction methods of binding affinity prediction models for protein design.

NetTCR-2.1: Lessons and guidance on how to develop models for TCR specificity predictions

T cell receptors (TCR) define the specificity of T cells and are responsible for their interaction with peptide antigen targets presented in complex with major histocompatibility complex (MHC) molecules. Understanding the rules underlying this interaction hence forms the foundation for our understanding of basic adaptive immunology. Over the last decade, efforts have been dedicated to developing assays for high throughput identification of peptide-specific TCRs. Based on such data, several computational methods have been proposed for predicting the TCR-pMHC interaction. The general conclusion from these studies is that the prediction of TCR interactions with MHC-peptide complexes remains highly challenging. Several reasons form the basis for this including scarcity and quality of data, and ill-defined modeling objectives imposed by the high redundancy of the available data. In this work, we propose a framework for dealing with this redundancy, allowing us to address essential questions related to the modeling of TCR specificity including the use of peptide- versus pan-specific models, how to best define negative data, and the performance impact of integrating of CDR1 and 2 loops. Further, we illustrate how and why it is strongly recommended to include simple similarity-based modeling approaches when validating an improved predictive power of machine learning models, and that such validation should include a performance evaluation as a function of "distance" to the training data, to quantify the potential for generalization of the proposed model. The conclusion of the work is that, given current data, TCR specificity is best modeled using peptide-specific approaches, integrating information from all 6 CDR loops, and with negative data constructed from a combination of true and mislabeled negatives. Comparing such machine learning models to similarity-based approaches demonstrated an increased performance gain of the former as the "distance" to the training data was increased; thus demonstrating an improved generalization ability of the machine learning-based approaches. We believe these results demonstrate that the outlined modeling framework and proposed evaluation strategy form a solid basis for investigating the modeling of TCR specificities and that adhering to such a framework will allow for faster progress within the field. The final devolved model, NetTCR-2.1, is available at https://services.healthtech.dtu.dk/service.php?NetTCR-2.1.

Deep learning reveals predictive sequence concepts within immune repertoires to immunotherapy

T cell receptor (TCR) sequencing has been used to characterize the immune response to cancer. However, most analyses have been restricted to quantitative measures such as clonality that do not leverage the complementarity-determining region 3 (CDR3) sequence. We use DeepTCR, a framework of deep learning algorithms, to reveal sequence concepts that are predictive of response to immunotherapy. We demonstrate that DeepTCR can predict response and use the model to infer the antigenic specificities of the predictive signature and their unique dynamics during therapy. The predictive signature of nonresponse is associated with high frequencies of TCRs predicted to recognize tumor-specific antigens, and these tumor-specific TCRs undergo a higher degree of dynamic changes on therapy in nonresponders versus responders. These results are consistent with a biological model where the hallmark of nonresponders is an accumulation of tumor-specific T cells that undergo turnover on therapy, possibly because of the dysfunctional state of these T cells in nonresponders.

ATM-TCR: TCR-Epitope Binding Affinity Prediction Using a Multi-Head Self-Attention Model

TCR-epitope pair binding is the key component for T cell regulation. The ability to predict whether a given pair binds is fundamental to understanding the underlying biology of the binding mechanism as well as developing T-cell mediated immunotherapy approaches. The advent of large-scale public databases containing TCR-epitope binding pairs enabled the recent development of computational prediction methods for TCR-epitope binding. However, the number of epitopes reported along with binding TCRs is far too small, resulting in poor out-of-sample performance for unseen epitopes. In order to address this issue, we present our model ATM-TCR which uses a multi-head self-attention mechanism to capture biological contextual information and improve generalization performance. Additionally, we present a novel application of the attention map from our model to improve out-of-sample performance by demonstrating on recent SARS-CoV-2 data.

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.

DECODE: a computational pipeline to discover T cell receptor binding rules

Motivation

Understanding the mechanisms underlying T cell receptor (TCR) binding is of fundamental importance to understanding adaptive immune responses. A better understanding of the biochemical rules governing TCR binding can be used, e.g. to guide the design of more powerful and safer T cell-based therapies. Advances in repertoire sequencing technologies have made available millions of TCR sequences. Data abundance has, in turn, fueled the development of many computational models to predict the binding properties of TCRs from their sequences. Unfortunately, while many of these works have made great strides toward predicting TCR specificity using machine learning, the black-box nature of these models has resulted in a limited understanding of the rules that govern the binding of a TCR and an epitope.

Results

We present an easy-to-use and customizable computational pipeline, DECODE, to extract the binding rules from any black-box model designed to predict the TCR-epitope binding. DECODE offers a range of analytical and visualization tools to guide the user in the extraction of such rules. We demonstrate our pipeline on a recently published TCR-binding prediction model, TITAN, and show how to use the provided metrics to assess the quality of the computed rules. In conclusion, DECODE can lead to a better understanding of the sequence motifs that underlie TCR binding. Our pipeline can facilitate the investigation of current immunotherapeutic challenges, such as cross-reactive events due to off-target TCR binding.

Availability and implementation

Code is available publicly at https://github.com/phineasng/DECODE.

Supplementary information

Supplementary data are available at Bioinformatics online.

A Deep Learning Model for Accurate Diagnosis of Infection Using Antibody Repertoires

The adaptive immune receptor repertoire consists of the entire set of an individual's BCRs and TCRs and is believed to contain a record of prior immune responses and the potential for future immunity. Analyses of TCR repertoires via deep learning (DL) methods have successfully diagnosed cancers and infectious diseases, including coronavirus disease 2019. However, few studies have used DL to analyze BCR repertoires. In this study, we collected IgG H chain Ab repertoires from 276 healthy control subjects and 326 patients with various infections. We then extracted a comprehensive feature set consisting of 10 subsets of repertoire-level features and 160 sequence-level features and tested whether these features can distinguish between infected individuals and healthy control subjects. Finally, we developed an ensemble DL model, namely, DL method for infection diagnosis (https://github.com/chenyuan0510/DeepID), and used this model to differentiate between the infected and healthy individuals. Four subsets of repertoire-level features and four sequence-level features were selected because of their excellent predictive performance. The DL method for infection diagnosis outperformed traditional machine learning methods in distinguishing between healthy and infected samples (area under the curve = 0.9883) and achieved a multiclassification accuracy of 0.9104. We also observed differences between the healthy and infected groups in V genes usage, clonal expansion, the complexity of reads within clone, the physical properties in the α region, and the local flexibility of the CDR3 amino acid sequence. Our results suggest that the Ab repertoire is a promising biomarker for the diagnosis of various infections.

Next-generation sequencing: unraveling genetic mechanisms that shape cancer immunotherapy efficacy

Immunity is governed by fundamental genetic processes. These processes shape the nature of immune cells and set the rules that dictate the myriad complex cellular interactions that power immune systems. Everything from the generation of T cell receptors and antibodies, control of epitope presentation, and recognition of pathogens by the immunoediting of cancer cells is, in large part, made possible by core genetic mechanisms and the cellular machinery that they encode. In the last decade, next-generation sequencing has been used to dissect the complexities of cancer immunity with potent effect. Sequencing of exomes and genomes has begun to reveal how the immune system recognizes “foreign” entities and distinguishes self from non-self, especially in the setting of cancer. High-throughput analyses of transcriptomes have revealed deep insights into how the tumor microenvironment affects immunotherapy efficacy. In this Review, we discuss how high-throughput sequencing has added to our understanding of how immune systems interact with cancer cells and how cancer immunotherapies work.

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.

Lupus Susceptibility Loci Predispose Mice to Clonal Lymphocytic Responses and Myeloid Expansion

Lupus susceptibility results from the combined effects of numerous genetic loci, but the contribution of these loci to disease pathogenesis has been difficult to study due to the large cellular heterogeneity of the autoimmune immune response. We performed single-cell RNA, BCR, and TCR sequencing of splenocytes from mice with multiple polymorphic lupus susceptibility loci. We not only observed lymphocyte and myeloid expansion, but we also characterized changes in subset frequencies and gene expression, such as decreased CD8 and marginal zone B cells and increased Fcrl5- and Cd5l-expressing macrophages. Clonotypic analyses revealed expansion of B and CD4 clones, and TCR repertoires from lupus-prone mice were distinguishable by algorithmic specificity prediction and unsupervised machine learning classification. Myeloid differential gene expression, metabolism, and altered ligand-receptor interaction were associated with decreased Ag presentation. This dataset provides novel mechanistic insight into the pathophysiology of a spontaneous model of lupus, highlighting potential therapeutic targets for autoantibody-mediated disease.

Utility of Bulk T-Cell Receptor Repertoire Sequencing Analysis in Understanding Immune Responses to COVID-19

Measuring immunity to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the causative agent of coronavirus disease 19 (COVID-19), can rely on antibodies, reactive T cells and other factors, with T-cell-mediated responses appearing to have greater sensitivity and longevity. Because each T cell carries an essentially unique nucleic acid sequence for its T-cell receptor (TCR), we can interrogate sequence data derived from DNA or RNA to assess aspects of the immune response. This review deals with the utility of bulk, rather than single-cell, sequencing of TCR repertoires, considering the importance of study design, in terms of cohort selection, laboratory methods and analysis. The advances in understanding SARS-CoV-2 immunity that have resulted from bulk TCR repertoire sequencing are also be discussed. The complexity of sequencing data obtained by bulk repertoire sequencing makes analysis challenging, but simple descriptive analyses, clonal analysis, searches for specific sequences associated with immune responses to SARS-CoV-2, motif-based analyses, and machine learning approaches have all been applied. TCR repertoire sequencing has demonstrated early expansion followed by contraction of SARS-CoV-2-specific clonotypes, during active infection. Maintenance of TCR repertoire diversity, including the maintenance of diversity of anti-SARS-CoV-2 response, predicts a favourable outcome. TCR repertoire narrowing in severe COVID-19 is most likely a consequence of COVID-19-associated lymphopenia. It has been possible to follow clonotypic sequences longitudinally, which has been particularly valuable for clonotypes known to be associated with SARS-CoV-2 peptide/MHC tetramer binding or with SARS-CoV-2 peptide-induced cytokine responses. Closely related clonotypes to these previously identified sequences have been shown to respond with similar kinetics during infection. A possible superantigen-like effect of the SARS-CoV-2 spike protein has been identified, by means of observing V-segment skewing in patients with severe COVID-19, together with structural modelling. Such a superantigen-like activity, which is apparently absent from other coronaviruses, may be the basis of multisystem inflammatory syndrome and cytokine storms in COVID-19. Bulk TCR repertoire sequencing has proven to be a useful and cost-effective approach to understanding interactions between SARS-CoV-2 and the human host, with the potential to inform the design of therapeutics and vaccines, as well as to provide invaluable pathogenetic and epidemiological insights.

TCR-sequencing in cancer and autoimmunity: barcodes and beyond

The T cell receptor (TCR) endows T cells with antigen specificity and is central to nearly all aspects of T cell function. Each naïve T cell has a unique TCR sequence that is stably maintained during cell division. In this way, the TCR serves as a molecular barcode that tracks processes such as migration, differentiation, and proliferation of T cells. Recent technological advances have enabled sequencing of the TCR from single cells alongside deep molecular phenotypes on an unprecedented scale. In this review, we discuss strengths and limitations of TCR sequences as molecular barcodes and their application to study immune responses following Programmed Death-1 (PD-1) blockade in cancer. Additionally, we consider applications of TCR data beyond use as a barcode.

Bystander T cells in cancer immunology and therapy

Cancer-specific T cells are required for effective anti-cancer immunity and have a central role in cancer immunotherapy. However, emerging evidence suggests that only a small fraction of tumor-infiltrating T cells are cancer specific, and T cells that recognize cancer-unrelated antigens (so-called 'bystanders') are abundant. Although the role of cancer-specific T cells in anti-cancer immunity has been well established, the implications of bystander T cells in tumors are only beginning to be understood. It is becoming increasingly clear that bystander T cells are not a homogeneous group of cells but, instead, they differ in their specificities, their activation states and effector functions. In this Perspective, we discuss recent studies of bystander T cells in tumors, including experimental and computational approaches that enable their identification and functional analysis and viewpoints on how these insights could be used to develop new therapeutic approaches for cancer immunotherapy.

Predicting antibody binders and generating synthetic antibodies using deep learning

The antibody drug field has continually sought improvements to methods for candidate discovery and engineering. Historically, most such methods have been laboratory-based, but informatics methods have recently started to make an impact. Deep learning, a subfield of machine learning, is rapidly gaining prominence in the biomedical research. Recent advances in microfluidics technologies and next-generation sequencing have not only revolutionized therapeutic antibody discovery, but also contributed to a vast amount of antibody repertoire sequencing data, providing opportunities for deep learning-based applications. Previously, we used microfluidics, yeast display, and deep sequencing to generate a panel of binder and non-binder antibody sequences to the cancer immunotherapy targets PD-1 and CTLA-4. Here we encoded the antibody light and heavy chain complementarity-determining regions (CDR3s) into antibody images, then built and trained convolutional neural network models to classify binders and non-binders. To improve model interpretability, we performed in silico mutagenesis to identify CDR3 residues that were important for binder classification. We further built generative deep learning models using generative adversarial network models to produce synthetic antibodies against PD-1 and CTLA-4. Our models generated variable length CDR3 sequences that resemble real sequences. Overall, our study demonstrates that deep learning methods can be leveraged to mine and learn patterns in antibody sequences, offering insights into antibody engineering, optimization, and discovery.

Heavy chain sequence-based classifier for the specificity of human antibodies

Antibodies specifically bind to antigens and are an essential part of the immune system. Hence, antibodies are powerful tools in research and diagnostics. High-throughput sequencing technologies have promoted comprehensive profiling of the immune repertoire, which has resulted in large amounts of antibody sequences that remain to be further analyzed. In this study, antibodies were downloaded from IMGT/LIGM-DB and Sequence Read Archive databases. Contributing features from antibody heavy chains were formulated as numerical inputs and fed into an ensemble machine learning classifier to classify the antigen specificity of six classes of antibodies, namely anti-HIV-1, anti-influenza virus, anti-pneumococcal polysaccharide, anti-citrullinated protein, anti-tetanus toxoid and anti-hepatitis B virus. The classifier was validated using cross-validation and a testing dataset. The ensemble classifier achieved a macro-average area under the receiver operating characteristic curve (AUC) of 0.9246 from the 10-fold cross-validation, and 0.9264 for the testing dataset. Among the contributing features, the contribution of the complementarity-determining regions was 53.1% and that of framework regions was 46.9%, and the amino acid mutation rates occupied the first and second ranks among the top five contributing features. The classifier and insights provided in this study could promote the mechanistic study, isolation and utilization of potential therapeutic antibodies.

Using the T Cell Receptor as a Biomarker in Type 1 Diabetes

T cell receptors (TCRs) are unique markers that define antigen specificity for a given T cell. With the evolution of sequencing and computational analysis technologies, TCRs are now prime candidates for the development of next-generation non-cell based T cell biomarkers, which provide a surrogate measure to assess the presence of antigen-specific T cells. Type 1 diabetes (T1D), the immune-mediated form of diabetes, is a prototypical organ specific autoimmune disease in which T cells play a pivotal role in targeting pancreatic insulin-producing beta cells. While the disease is now predictable by measuring autoantibodies in the peripheral blood directed to beta cell proteins, there is an urgent need to develop T cell markers that recapitulate T cell activity in the pancreas and can be a measure of disease activity. This review focuses on the potential and challenges of developing TCR biomarkers for T1D. We summarize current knowledge about TCR repertoires and clonotypes specific for T1D and discuss challenges that are unique for autoimmune diabetes. Ultimately, the integration of large TCR datasets produced from individuals with and without T1D along with computational 'big data' analysis will facilitate the development of TCRs as potentially powerful biomarkers in the development of T1D.

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.

Artificial intelligence: A powerful paradigm for scientific research

Artificial intelligence (AI) coupled with promising machine learning (ML) techniques well known from computer science is broadly affecting many aspects of various fields including science and technology, industry, and even our day-to-day life. The ML techniques have been developed to analyze high-throughput data with a view to obtaining useful insights, categorizing, predicting, and making evidence-based decisions in novel ways, which will promote the growth of novel applications and fuel the sustainable booming of AI. This paper undertakes a comprehensive survey on the development and application of AI in different aspects of fundamental sciences, including information science, mathematics, medical science, materials science, geoscience, life science, physics, and chemistry. The challenges that each discipline of science meets, and the potentials of AI techniques to handle these challenges, are discussed in detail. Moreover, we shed light on new research trends entailing the integration of AI into each scientific discipline. The aim of this paper is to provide a broad research guideline on fundamental sciences with potential infusion of AI, to help motivate researchers to deeply understand the state-of-the-art applications of AI-based fundamental sciences, and thereby to help promote the continuous development of these fundamental sciences.

Follicular T cells are clonally and transcriptionally distinct in B cell-driven mouse autoimmune disease

Pathogenic autoantibodies contribute to tissue damage and clinical decline in autoimmune disease. Follicular T cells are central regulators of germinal centers, although their contribution to autoantibody-mediated disease remains unclear. Here we perform single cell RNA and T cell receptor (TCR) sequencing of follicular T cells in a mouse model of autoantibody-mediated disease, allowing for analyses of paired transcriptomes and unbiased TCRαβ repertoires at single cell resolution. A minority of clonotypes are preferentially shared amongst autoimmune follicular T cells and clonotypic expansion is associated with differential gene signatures in autoimmune disease. Antigen prediction using algorithmic and machine learning approaches indicates convergence towards shared specificities between non-autoimmune and autoimmune follicular T cells. However, differential autoimmune transcriptional signatures are preserved even amongst follicular T cells with shared predicted specificities. These results demonstrate that follicular T cells are phenotypically distinct in B cell-driven autoimmune disease, providing potential therapeutic targets to modulate autoantibody development.

Comparative evaluation of T-cell receptors in experimental glioma-draining lymph nodes

Background

Glioblastomas, the most common primary malignant brain tumors, are considered immunologically cold malignancies due to growth in an immune sanctuary site. While peptide vaccines have shown to generate intra-tumoral antigen-specific T cells, the identification of these tumor-specific T cells is challenging and requires detailed analyses of tumor tissue. Several studies have shown that CNS antigens may be transported via lymphatic drainage to cervical lymph nodes, where antigen-specific T-cell responses can be generated. Therefore, we investigated whether glioma-draining lymph nodes (TDLN) may constitute a reservoir of tumor-reactive T cells.

Methods

We addressed our hypothesis by flow cytometric analyses of chicken ovalbumin (OVA)-specific CD8⁺ T cells as well as T-cell receptor beta (TCRβ) next-generation-sequencing (TCRβ-NGS) of T cells from tumor tissue, TDLN, spleen, and inguinal lymph nodes harvested from experimental mouse GL261 glioma models.

Results

Longitudinal dextramer-based assessment of specific CD8⁺ T cells from TDLN did not show tumor model antigen reactivity. Unbiased immunogenomic analysis revealed a low overlap of TCRβ sequences from glioma-infiltrating CD8⁺ T cells between mice. Enrichment scores, calculated by the ratio of productive frequencies of the different TCRβ-CDR3 amino-acid (aa) rearrangements of CD8⁺ T cells derived from tumor, TDLN, inguinal lymph nodes, and spleen demonstrated a higher proportion of tumor-associated TCR in the spleen compared to TDLN.

Conclusions

In experimental glioblastoma, our data did not provide evidence that glioma-draining cervical lymph nodes are a robust reservoir for spontaneous glioma-specific T cells highlighting the requirement for detailed analyses of glioma-infiltrating T cells for the discovery of tumor-specific TCR.

Single-Cell Analysis Using Machine Learning Techniques and Its Application to Medical Research

In recent years, the diversity of cancer cells in tumor tissues as a result of intratumor heterogeneity has attracted attention. In particular, the development of single-cell analysis technology has made a significant contribution to the field; technologies that are centered on single-cell RNA sequencing (scRNA-seq) have been reported to analyze cancer constituent cells, identify cell groups responsible for therapeutic resistance, and analyze gene signatures of resistant cell groups. However, although single-cell analysis is a powerful tool, various issues have been reported, including batch effects and transcriptional noise due to gene expression variation and mRNA degradation. To overcome these issues, machine learning techniques are currently being introduced for single-cell analysis, and promising results are being reported. In addition, machine learning has also been used in various ways for single-cell analysis, such as single-cell assay of transposase accessible chromatin sequencing (ATAC-seq), chromatin immunoprecipitation sequencing (ChIP-seq) analysis, and multi-omics analysis; thus, it contributes to a deeper understanding of the characteristics of human diseases, especially cancer, and supports clinical applications. In this review, we present a comprehensive introduction to the implementation of machine learning techniques in medical research for single-cell analysis, and discuss their usefulness and future potential.

Predicting T Cell Receptor Antigen Specificity From Structural Features Derived From Homology Models of Receptor-Peptide-Major Histocompatibility Complexes

The physical interaction between the T cell receptor (TCR) and its cognate antigen causes T cells to activate and participate in the immune response. Understanding this physical interaction is important in predicting TCR binding to a target epitope, as well as potential cross-reactivity. Here, we propose a way of collecting informative features of the binding interface from homology models of T cell receptor-peptide-major histocompatibility complex (TCR-pMHC) complexes. The information collected from these structures is sufficient to discriminate binding from non-binding TCR-pMHC pairs in multiple independent datasets. The classifier is limited by the number of crystal structures available for the homology modelling and by the size of the training set. However, the classifier shows comparable performance to sequence-based classifiers requiring much larger training sets.

Quantifying T Cell Cross-Reactivity: Influenza and Coronaviruses

If viral strains are sufficiently similar in their immunodominant epitopes, then populations of cross-reactive T cells may be boosted by exposure to one strain and provide protection against infection by another at a later date. This type of pre-existing immunity may be important in the adaptive immune response to influenza and to coronaviruses. Patterns of recognition of epitopes by T cell clonotypes (a set of cells sharing the same T cell receptor) are represented as edges on a bipartite network. We describe different methods of constructing bipartite networks that exhibit cross-reactivity, and the dynamics of the T cell repertoire in conditions of homeostasis, infection and re-infection. Cross-reactivity may arise simply by chance, or because immunodominant epitopes of different strains are structurally similar. We introduce a circular space of epitopes, so that T cell cross-reactivity is a quantitative measure of the overlap between clonotypes that recognize similar (that is, close in epitope space) epitopes.

Probing T-cell response by sequence-based probabilistic modeling

With the increasing ability to use high-throughput next-generation sequencing to quantify the diversity of the human T cell receptor (TCR) repertoire, the ability to use TCR sequences to infer antigen-specificity could greatly aid potential diagnostics and therapeutics. Here, we use a machine-learning approach known as Restricted Boltzmann Machine to develop a sequence-based inference approach to identify antigen-specific TCRs. Our approach combines probabilistic models of TCR sequences with clone abundance information to extract TCR sequence motifs central to an antigen-specific response. We use this model to identify patient personalized TCR motifs that respond to individual tumor and infectious disease antigens, and to accurately discriminate specific from non-specific responses. Furthermore, the hidden structure of the model results in an interpretable representation space where TCRs responding to the same antigen cluster, correctly discriminating the response of TCR to different viral epitopes. The model can be used to identify condition specific responding TCRs. We focus on the examples of TCRs reactive to candidate neoantigens and selected epitopes in experiments of stimulated TCR clone expansion.

The methods and advances of adaptive immune receptors repertoire sequencing

The adaptive immune response is a powerful tool, capable of recognizing, binding to, and neutralizing a vast number of internal and external threats via T or B lymphatic receptors with widespread sets of antigen specificities. The emergence of high-throughput sequencing technology and bioinformatics provides opportunities for research in the fields of life sciences and medicine. The analysis and annotation for immune repertoire data can reveal biologically meaningful information, including immune prediction, target antigens, and effective evaluation. Continuous improvements of the immunological repertoire sequencing methods and analysis tools will help to minimize the experimental and calculation errors and realize the immunological information to meet the clinical requirements. That said, the clinical application of adaptive immune repertoire sequencing requires appropriate experimental methods and standard analytical tools. At the population cell level, we can acquire the overview of cell groups, but the information about a single cell is not obtained accurately. The information that is ignored may be crucial for understanding the heterogeneity of each cell, gene expression and drug response. The combination of high-throughput sequencing and single-cell technology allows us to obtain single-cell information with low-cost and high-throughput. In this review, we summarized the current methods and progress in this area.

Antigen-Specific Treg Therapy in Type 1 Diabetes - Challenges and Opportunities

Regulatory T cells (Tregs) are key mediators of peripheral self-tolerance and alterations in their frequencies, stability, and function have been linked to autoimmunity. The antigen-specific induction of Tregs is a long-envisioned goal for the treatment of autoimmune diseases given reduced side effects compared to general immunosuppressive therapies. However, the translation of antigen-specific Treg inducing therapies for the treatment or prevention of autoimmune diseases into the clinic remains challenging. In this mini review, we will discuss promising results for antigen-specific Treg therapies in allergy and specific challenges for such therapies in autoimmune diseases, with a focus on type 1 diabetes (T1D). We will furthermore discuss opportunities for antigen-specific Treg therapies in T1D, including combinatorial strategies and tissue-specific Treg targeting. Specifically, we will highlight recent advances in miRNA-targeting as a means to foster Tregs in autoimmunity. Additionally, we will discuss advances and perspectives of computational strategies for the detailed analysis of tissue-specific Tregs on the single-cell level.

Current challenges for unseen-epitope TCR interaction prediction and a new perspective derived from image classification

The prediction of epitope recognition by T-cell receptors (TCRs) has seen many advancements in recent years, with several methods now available that can predict recognition for a specific set of epitopes. However, the generic case of evaluating all possible TCR-epitope pairs remains challenging, mainly due to the high diversity of the interacting sequences and the limited amount of currently available training data. In this work, we provide an overview of the current state of this unsolved problem. First, we examine appropriate validation strategies to accurately assess the generalization performance of generic TCR-epitope recognition models when applied to both seen and unseen epitopes. In addition, we present a novel feature representation approach, which we call ImRex (interaction map recognition). This approach is based on the pairwise combination of physicochemical properties of the individual amino acids in the CDR3 and epitope sequences, which provides a convolutional neural network with the combined representation of both sequences. Lastly, we highlight various challenges that are specific to TCR-epitope data and that can adversely affect model performance. These include the issue of selecting negative data, the imbalanced epitope distribution of curated TCR-epitope datasets and the potential exchangeability of TCR alpha and beta chains. Our results indicate that while extrapolation to unseen epitopes remains a difficult challenge, ImRex makes this feasible for a subset of epitopes that are not too dissimilar from the training data. We show that appropriate feature engineering methods and rigorous benchmark standards are required to create and validate TCR-epitope predictive models.

TITAN: T-cell receptor specificity prediction with bimodal attention networks

MOTIVATION

The activity of the adaptive immune system is governed by T-cells and their specific T-cell receptors (TCR), which selectively recognize foreign antigens. Recent advances in experimental techniques have enabled sequencing of TCRs and their antigenic targets (epitopes), allowing to research the missing link between TCR sequence and epitope binding specificity. Scarcity of data and a large sequence space make this task challenging, and to date only models limited to a small set of epitopes have achieved good performance. Here, we establish a k-nearest-neighbor (K-NN) classifier as a strong baseline and then propose Tcr epITope bimodal Attention Networks (TITAN), a bimodal neural network that explicitly encodes both TCR sequences and epitopes to enable the independent study of generalization capabilities to unseen TCRs and/or epitopes.

RESULTS

By encoding epitopes at the atomic level with SMILES sequences, we leverage transfer learning and data augmentation to enrich the input data space and boost performance. TITAN achieves high performance in the prediction of specificity of unseen TCRs (ROC-AUC 0.87 in 10-fold CV) and surpasses the results of the current state-of-the-art (ImRex) by a large margin. Notably, our Levenshtein-based K-NN classifier also exhibits competitive performance on unseen TCRs. While the generalization to unseen epitopes remains challenging, we report two major breakthroughs. First, by dissecting the attention heatmaps, we demonstrate that the sparsity of available epitope data favors an implicit treatment of epitopes as classes. This may be a general problem that limits unseen epitope performance for sufficiently complex models. Second, we show that TITAN nevertheless exhibits significantly improved performance on unseen epitopes and is capable of focusing attention on chemically meaningful molecular structures.

AVAILABILITY AND IMPLEMENTATION

The code as well as the dataset used in this study is publicly available at https://github.com/PaccMann/TITAN.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Deep learning identifies antigenic determinants of severe SARS-CoV-2 infection within T-cell repertoires

SARS-CoV-2 infection is characterized by a highly variable clinical course with patients experiencing asymptomatic infection all the way to requiring critical care support. This variation in clinical course has led physicians and scientists to study factors that may predispose certain individuals to more severe clinical presentations in hopes of either identifying these individuals early in their illness or improving their medical management. We sought to understand immunogenomic differences that may result in varied clinical outcomes through analysis of T-cell receptor sequencing (TCR-Seq) data in the open access ImmuneCODE database. We identified two cohorts within the database that had clinical outcomes data reflecting severity of illness and utilized DeepTCR, a multiple-instance deep learning repertoire classifier, to predict patients with severe SARS-CoV-2 infection from their repertoire sequencing. We demonstrate that patients with severe infection have repertoires with higher T-cell responses associated with SARS-CoV-2 epitopes and identify the epitopes that result in these responses. Our results provide evidence that the highly variable clinical course seen in SARS-CoV-2 infection is associated to certain antigen-specific responses.

Autoencoder based local T cell repertoire density can be used to classify samples and T cell receptors

Recent advances in T cell repertoire (TCR) sequencing allow for the characterization of repertoire properties, as well as the frequency and sharing of specific TCR. However, there is no efficient measure for the local density of a given TCR. TCRs are often described either through their Complementary Determining region 3 (CDR3) sequences, or theirV/J usage, or their clone size. We here show that the local repertoire density can be estimated using a combined representation of these components through distance conserving autoencoders and Kernel Density Estimates (KDE). We present ELATE-an Encoder-based LocAl Tcr dEnsity and show that the resulting density of a sample can be used as a novel measure to study repertoire properties. The cross-density between two samples can be used as a similarity matrix to fully characterize samples from the same host. Finally, the same projection in combination with machine learning algorithms can be used to predict TCR-peptide binding through the local density of known TCRs binding a specific target.

Biological controls for standardization and interpretation of adaptive immune receptor repertoire profiling

Use of adaptive immune receptor repertoire sequencing (AIRR-seq) has become widespread, providing new insights into the immune system with potential broad clinical and diagnostic applications. However, like many high-throughput technologies, it comes with several problems, and the AIRR Community was established to understand and help solve them. We, the AIRR Community's Biological Resources Working Group, have surveyed scientists about the need for standards and controls in generating and annotating AIRR-seq data. Here, we review the current status of AIRR-seq, provide the results of our survey, and based on them, offer recommendations for developing AIRR-seq standards and controls, including future work.

Predicting recognition between T cell receptors and epitopes with TCRGP

Adaptive immune system uses T cell receptors (TCRs) to recognize pathogens and to consequently initiate immune responses. TCRs can be sequenced from individuals and methods analyzing the specificity of the TCRs can help us better understand individuals' immune status in different disorders. For this task, we have developed TCRGP, a novel Gaussian process method that predicts if TCRs recognize specified epitopes. TCRGP can utilize the amino acid sequences of the complementarity determining regions (CDRs) from TCRα and TCRβ chains and learn which CDRs are important in recognizing different epitopes. Our comprehensive evaluation with epitope-specific TCR sequencing data shows that TCRGP achieves on average higher prediction accuracy in terms of AUROC score than existing state-of-the-art methods in epitope-specificity predictions. We also propose a novel analysis approach for combined single-cell RNA and TCRαβ (scRNA+TCRαβ) sequencing data by quantifying epitope-specific TCRs with TCRGP and identify HBV-epitope specific T cells and their transcriptomic states in hepatocellular carcinoma patients.