Biology subjects

Reddy, S. T.

Publications and source records attributed to Reddy, S. T..

At least 37 records · Page 2Linked to original sources

bioRxiv2024

Protein language model pseudolikelihoods capture features of in vivo B cell selection and evolution

B cell selection and evolution play crucial roles in dictating successful immune responses. Recent advancements in sequencing technologies and deep-learning strategies have paved the way for generating and exploiting an ever-growing wealth of antibody repertoire data. The self-supervised nature of protein language models (PLMs) has demonstrated the ability to learn complex representations of antibody sequences and has been leveraged for a wide range of applications including diagnostics, structural modeling, and antigen-specificity predictions. PLM-derived likelihoods have been used to improve antibody affinities in vitro, raising the question of whether PLMs can capture and predict features of B cell selection in vivo. Here, we explore how general and antibody-specific PLM-generated sequence pseudolikelihoods (SPs) relate to features of in vivo B cell selection such as expansion, isotype usage, and somatic hypermutation (SHM) at single-cell resolution. Our results demonstrate that the type of PLM and the region of the antibody input sequence significantly affect the generated SP. Contrary to previous in vitro reports, we observe a negative correlation between SPs and binding affinity, whereas repertoire features such as SHM and isotype usage were strongly correlated with SPs. By constructing evolutionary lineage trees of B cell clones from human and mouse repertoires, we observe that SHMs are routinely among the most likely mutations suggested by PLMs and that mutating residues have lower absolute likelihoods than conserved residues. Our findings highlight the potential of PLMs to predict features of antibody selection and further suggest their potential to assist in antibody discovery and engineering. Key points- In contrast to previous in vitro work (Hie et al., 2024), we observe a negative correlation between PLM-generated SP and binding affinity. This contrast can be explained by the inherent antibody germline bias posed by PLM training data and the difference between in vivo and in vitro settings. - Our findings also reveal a considerable correlation between SPs and repertoire features such as the V-gene family, isotype, and the amount of SHM. Moreover, labeled antigen-binding data suggested that SP is consistent with antigen-specificity and binding affinity. - By reconstructing B cell lineage evolutionary trajectories, we detected predictable features of SHM using PLMs. We observe that SHMs are routinely among the most likely mutations suggested by PLMs and that mutating residues have lower absolute likelihoods than conserved residues. - We demonstrate that the region of antibody sequence (CDR3 or full V(D)J) provided as input to the model, as well as the type of PLM used, influence the resulting SPs.

bioinformatics↗

bioRxiv2024

Clonally expanded IgG antibody-secreting cells preferentially target influenza nucleoprotein following homologous and heterologous infections

Infection with influenza virus remains a significant global health concern due to its ability to acquire mutations at key antigenic sites to escape antibody recognition. While germinal center (GC) and memory B cells have been well studied following influenza infection, the clonal dynamics of antibody secreting cells (ASCs), particularly those within the bone marrow (BM) niche that are responsible for serum immune protection, remain poorly understood. Here, we combine single-cell RNA (scRNA) and B cell receptor (BCR) sequencing to characterize individual ASCs following various Influenza exposure histories. We find that BM repertories are populated by highly expanded and class-switched ASCs following Influenza infection with similar transcriptional and repertoire characteristics regardless of homologous or heterologous infection histories. By combining single-cell analysis with monoclonal antibody expression and characterization, we find that a large proportion of the expanded IgG-, but not IgA-, ASC repertoire demonstrates specificity to influenza nucleoprotein (NP). Together, our data reveal the complex relationship between BM ASC repertoires, mucosal humoral immune responses, and BCR antigen specificity during influenza infection.

immunology↗

bioRxiv2024

Modeling CAR Response at the Single-Cell Level Using Conditional Optimal Transport

Chimeric Antigen Receptor (CAR) T cell therapy is a promising area of cancer immunotherapy. However, many challenges such as loss of persistence, T cell exhaustion, and therapy associated toxicities hamper further advancement of CAR T cell therapy. Therefore, recent efforts have focused on designing improved CARs that show better therapeutic characteristics. However, it is unfeasible to test all CAR variants in lab-based assays as CARs consist of multiple intracellular signalling domains. This results in over 100000 possible variants. We leverage computational modeling to navigate this vast combinatorial space by learning the relationship between CAR design and T cell functionality, thereby proposing promising CAR T cell designs. CAR T cells expressing different variants can be viewed as cells that underwent different perturbations. Neural Optimal Transport is an upcoming field that can model single cell perturbations and predict unseen cells and conditions. In this work we leverage the conditional Monge Gap to model the response to CAR expression at a single-cell level and generate gene expression of cells that express an unseen CAR design. We show that CAR OT (CAROT) significantly outperforms the baseline for gene expression prediction for in-distribution CAR variants, with distinct gene expression patterns per CAR that capture biological characteristics. When predicting unseen CAR variants, we demonstrate promising results in terms of gene expression prediction and show the model learns gene expression patterns linked to domains in the training set. This work demonstrates that optimal transport may support discovery and development of new CAR T cell designs.

synthetic biology↗

bioRxiv2024

TCR clustering by contrastive learning on antigen specificity.

Effective clustering of T-cell receptor (TCR) sequences could be used to predict their antigen-specificities. TCRs with highly dissimilar sequences can bind to the same antigen, thus making their clustering into a common antigen group a central challenge. Here, we develop TouCAN, a method that relies on contrastive learning and pre-trained protein language models to perform TCR sequence clustering and antigen-specificity predictions. Following training, TouCAN demonstrates the ability to cluster highly dissimilar TCRs into common antigen groups. Additionally, TouCAN demonstrates TCR clustering performance and antigen-specificity predictions comparable to other leading methods in the field.

bioinformatics↗

bioRxiv2024

Synthetic coevolution reveals adaptive mutational trajectories of neutralizing antibodies and SARS-CoV-2

The Covid-19 pandemic showcases a coevolutionary race between the human immune system and SARS-CoV-2, mirroring the Red Queen hypothesis of evolutionary biology. The immune system generates neutralizing antibodies targeting the SARS-CoV-2 spike proteins receptor binding domain (RBD), crucial for host cell invasion, while the virus evolves to evade antibody recognition. Here, we establish a synthetic coevolution system combining high-throughput screening of antibody and RBD variant libraries with protein mutagenesis, surface display, and deep sequencing. Additionally, we train a protein language machine learning model that predicts antibody escape to RBD variants. Synthetic coevolution reveals antagonistic and compensatory mutational trajectories of neutralizing antibodies and SARS-CoV-2 variants, enhancing the understanding of this evolutionary conflict.

synthetic biology↗

bioRxiv2024

Dissecting the role of CAR signaling architectures on T cell activation and persistence using pooled screening and single-cell sequencing

Chimeric antigen receptor (CAR) T cells represent a promising approach for cancer treatment, yet challenges remain such as limited efficacy due to a lack of T cell persistence. Given its critical role in promoting and modulating T cell responses, it is crucial to understand how alterations in the CAR signaling architecture influence T cell function. Here, we designed a combinatorial CAR signaling domain library and performed repeated antigen stimulation assays, pooled screening and single-cell sequencing to investigate T-cell responses triggered by different CAR architectures. Parallel comparisons of CAR variants, at early, middle and late timepoints during chronic antigen stimulation systematically assessed the impact of modifying signaling domains on T cell activation and persistence. Our data reveal the predominant influence of membrane-proximal domains in driving T cell phenotype. Additionally, we highlight the critical role of CD40 costimulation in promoting potent and persistent T cell responses, followed by CTLA4, which induces a long-term cytotoxic phenotype. This work deepens the understanding of CAR T cell biology and may be used to guide the future engineering of CAR T cell therapies.

synthetic biology↗

bioRxiv2023

Deep learning-guided selection of antibody therapies with enhanced resistance to current and prospective SARS-CoV-2 Omicron variants

Most COVID-19 antibody therapies rely on binding the SARS-CoV-2 receptor binding domain (RBD). However, heavily mutated variants such as Omicron and its sublineages, which are characterized by an ever increasing number of mutations in the RBD, have rendered prior antibody therapies ineffective, leaving no clinically approved antibody treatments for SARS-CoV-2. Therefore, the capacity of therapeutic antibody candidates to bind and neutralize current and prospective SARS-CoV-2 variants is a critical factor for drug development. Here, we present a deep learning-guided approach to identify antibodies with enhanced resistance to SARS-CoV-2 evolution. We apply deep mutational learning (DML), a machine learning-guided protein engineering method to interrogate a massive sequence space of combinatorial RBD mutations and predict their impact on angiotensin-converting enzyme 2 (ACE2) binding and antibody escape. A high mutational distance library was constructed based on the full-length RBD of Omicron BA.1, which was experimentally screened for binding to the ACE2 receptor or neutralizing antibodies, followed by deep sequencing. The resulting data was used to train ensemble deep learning models that could accurately predict binding or escape for a panel of therapeutic antibody candidates targeting diverse RBD epitopes. Furthermore, antibody breadth was assessed by predicting binding or escape to synthetic lineages that represent millions of sequences generated using in silico evolution, revealing combinations with complementary and enhanced resistance to viral evolution. This deep learning approach may enable the design of next-generation antibody therapies that remain effective against future SARS-CoV-2 variants.

immunology↗

bioRxiv2023

Multimodal single-cell profiling of T cell specificity and reactivity in lung cancer

Adoptive transfer of autologous tumor-infiltrating lymphocyte T cells (TILs) offers one of the most promising approaches for cancer immunotherapy. However, high variability in patient responses highlight the need for an enhanced understanding of the transcriptional phenotypes of TILs and reactivity of their T cell receptors (TCR). Here, we employ single-cell multiomics approaches and TCR functional screening to investigate TILs from treatment-naive non-small cell lung cancer patients. This comprehensive analysis integrates scRNA-seq, scTCR-seq, and scATAC-seq, enabling a high-resolution examination of TILs within lung cancer tissue, as well as the adjacent non-tumor tissue. We apply a cellular functional screening platform to identify reactive TCRs that represent >1,000 TILs and have specificity towards a multitude of targets, including primary tumor cells, neoantigens, tumor-associated antigens, and viral antigens. Tumor-reactive TILs were primarily associated with dysfunctional phenotypes, whereas viral antigen-reactive TCRs were found in effector phenotype clusters. Key marker genes were identified and used to construct a tumor or viral reactivity score. Comparing clones shared in tumor and non-tumor tissue, a higher fraction of exhausted cells was observed in the tumor tissue, whereas non-tumor adjacent tissue possessed more effector cells, thus providing insight into potential sources for therapeutic T cells. Elucidating the specific T cell populations within TILs and their associated TCRs may support strategies to enhance the efficacy of TIL-based therapies. Graphical AbstractO_ST_ABSMultimodal single cell profiling and reactivity testing of TILsC_ST_ABS(A) CD8+ T cells of treatment naive non-small cell lung cancer patients and adjacent lung tissue were isolated by fluorescence-activated cell sorting (FACS) and were then subjected to scRNA-seq + scTCR-seq or scATAC-seq. (B) TCRs were functionally screened using a cellular platform (TnT cells) and target cells (tumor cells, antigen-pulsed antigen-presenting cells, PBMCs) by flow cytometry and deep sequencing. (C) scRNA-seq + scATAC-seq allowed trajectory inference of transcription factors and genes along pseudotime. (D) Gene scores for tumor- and virus-reactivity were developed by combining functional reactivity and transcriptomic profiling for each CD8+ T cell. (E) TIL scRNA-seq pre and post IL-2 treatment in tumor suspension displayed as alluvial plot shows change of clonal cell state composition. O_FIG O_LINKSMALLFIG WIDTH=196 HEIGHT=200 SRC="FIGDIR/small/560863v1_ufig1.gif" ALT="Figure 1"> View larger version (51K): org.highwire.dtl.DTLVardef@12fcc20org.highwire.dtl.DTLVardef@1027dccorg.highwire.dtl.DTLVardef@911711org.highwire.dtl.DTLVardef@18171e5_HPS_FORMAT_FIGEXP M_FIG C_FIG

immunology↗

bioRxiv2023

Meta Learning Improves Robustness and Performance in Machine Learning-Guided Protein Engineering

Machine learning-guided protein engineering continues to rapidly progress, however, collecting large, well-labeled data sets remains time and resource intensive. Directed evolution and protein engineering studies often require extensive experimental processes to eliminate noise and fully label high-throughput protein sequence-function data. Meta learning methods established in other fields (e.g. computer vision and natural language processing) have proven effective in learning from noisy data, given the availability of a small data set with trusted labels and thus could be applied for protein engineering. Here, we generate yeast display antibody mutagenesis libraries and screen them for target antigen binding followed by deep sequencing. Meta learning approaches are able to learn under high synthetic and experimental noise as well as in under labeled data settings, typically outperforming baselines significantly and often requiring a fraction of the training data. Thus, we demonstrate meta learning may expedite and improve machine learning-guided protein engineering. Availability and implementationThe code used in this study is publicly available at https://github.com/LSSI-ETH/meta-learning-for-protein-engineering. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=184 SRC="FIGDIR/small/526201v1_ufig1.gif" ALT="Figure 1"> View larger version (37K): org.highwire.dtl.DTLVardef@11d643forg.highwire.dtl.DTLVardef@a07828org.highwire.dtl.DTLVardef@130c8fdorg.highwire.dtl.DTLVardef@1db796d_HPS_FORMAT_FIGEXP M_FIG C_FIG

bioengineering↗

bioRxiv2022

A method for polyclonal antigen-specific T cell-targeted genome editing (TarGET) for adoptive cell therapy applications

Adoptive cell therapy of donor-derived, antigen-specific T cells expressing native T cell receptors (TCRs) is a powerful strategy to fight viral infections in immunocompromised patients. Determining the fate of T cells following patient infusion hinges on the ability to track them in vivo. While this is possible by genetic labeling of parent cells, the applicability of this approach has been limited by the non-specificity of the edited T cells. Here, we devised a method for CRISPR-targeted genome integration of a barcoded gene into Epstein-Barr virus-antigen-stimulated T cells and demonstrated its use for exclusively identifying expanded virus-specific cell lineages. Our method facilitated the enrichment of antigen-specific T cells, which then mediated improved cytotoxicity against EBV-transformed target cells. Single-cell and deep sequencing for lineage tracing revealed the expansion profile of specific T cell clones and their corresponding gene expression signature. This method has the potential to enhance the traceability and the monitoring capabilities during immunotherapeutic T cell regimens.

immunology↗

bioRxiv2022

ePlatypus: an ecosystem for computational analysis of immunogenomics data

The maturation of systems immunology methodologies requires novel and transparent computational frameworks capable of integrating diverse data modalities in a reproducible manner. Here, we present the ePlatypus computational immunology ecosystem for immunogenomics data analysis, with a focus on adaptive immune repertoires and single-cell sequencing. ePlatypus is a web-based platform and provides programming tutorials and an integrative database that elucidates selection patterns of adaptive immunity. Furthermore, the ecosystem links novel and established bioinformatics pipelines relevant for single-cell immune repertoires and other aspects of computational immunology such as predicting ligand-receptor interactions, structural modeling, simulations, machine learning, graph theory, pseudotime, spatial transcriptomics and phylogenetics. The ePlatypus ecosystem helps extract deeper insight in computational immunology and immunogenomics and promote open science. Accessibilityhttps://alexyermanos.github.io/Platypus/index.html

immunology↗

bioRxiv2022

Omicron sublineage BA.2.75.2 exhibits extensive escape from neutralising antibodies

Several sublineages of omicron have emerged with additional mutations that may afford further antibody evasion. Here, we characterise the sensitivity of emerging omicron sublineages BA.2.75.2, BA.4.6, and BA.2.10.4 to antibody-mediated neutralisation, and identify extensive escape by BA.2.75.2. BA.2.75.2 was resistant to neutralisation by Evusheld (tixagevimab + cilgavimab), but remained sensitive to bebtelovimab. In recent serum samples from blood donors in Stockholm, Sweden, BA.2.75.2 was neutralised, on average, at titers approximately 6.5-times lower than BA.5, making BA.2.75.2 the most neutralisation resistant variant evaluated to date. These data raise concerns that BA.2.75.2 may effectively evade humoral immunity in the population.

immunology↗

bioRxiv2022

Persistent virus-specific and clonally expanded antibody secreting cells respond to induced self antigen in the CNS

B cells contribute to the pathogenesis of both cellular- and humoral-mediated central nervous system (CNS) inflammatory diseases through a variety of mechanisms. In such conditions, B cells may enter the CNS parenchyma and contribute to local tissue destruction. It remains unexplored, however, how infection and autoimmunity drive transcriptional phenotypes, repertoire features, and antibody functionality. Here, we profiled B cells from the CNS of murine models of intracranial (i.c.) viral infections and autoimmunity. We identified a population of clonally expanded, antibody secreting cells (ASCs) that had undergone class-switch recombination and extensive somatic hypermutation following i.c. infection with attenuated lymphocytic choriomeningitis virus (rLCMV). Recombinant expression and characterisation of these antibodies revealed specificity to viral antigens (LCMV glycoprotein GP), correlating with ASC persistence in the brain weeks after resolved infection. Furthermore, these virus-specific ASCs upregulated proliferation and expansion programs in response to the conditional and transient induction of the LCMV GP as a neo-self antigen by astrocytes. This class-switched, clonally expanded, and mutated population persisted and was even more pronounced when peripheral B cells were depleted prior to autoantigen induction in the CNS. In contrast, the most expanded B cell clones in mice with persistent expression of LCMV GP in the CNS did not exhibit neo-self antigen specificity, potentially a consequence of local tolerance induction. Finally, a comparable population of clonally expanded, class-switched, proliferating ASCs was detected in the cerebrospinal fluid of multiple sclerosis patients. Taken together, our findings support the existence of B cells that populate the CNS and are capable of responding to locally encountered autoantigens. Graphical abstract O_FIG_DISPLAY_L [Figure 1] M_FIG_DISPLAY C_FIG_DISPLAY

immunology↗

bioRxiv2022

Evasion of neutralizing antibodies by Omicron sublineage BA.2.75

An emerging SARS-CoV-2 Omicron sublineage, BA.2.75, is increasing in frequency in India and has been detected in at least 15 countries as of 19 July 2022. Relative to BA.2, BA.2.75 carries nine additional mutations in spike. Here we report the sensitivity of the BA.2.75 spike to neutralization by a panel of clinically-relevant and pre-clinical monoclonal antibodies, as well as by serum from blood donated in Stockholm, Sweden, before and after the BA.1/BA.2 infection wave. BA.2.75 largely maintains sensitivity to bebtelovimab, despite a slight reduction in potency, and exhibits moderate susceptibility to tixagevimab and cilgavimab. For sera sampled both before and after the BA.1/BA.2 infection wave, BA.2.75 does not show significantly greater antibody evasion than the currently-dominating BA.5.

immunology↗

bioRxiv2022

Nucleotide augmentation for machine learning-guided protein engineering

Machine learning-guided protein engineering is a rapidly advancing field. Despite major experimental and computational advances however, collecting protein genotype (sequence) and phenotype (function) data remains time and resource intensive. As a result, the quality and quantity of training data is often a limiting factor in developing machine learning models. Data augmentation techniques have been successfully applied to the fields of computer vision and natural language processing, however, there is a lack of such augmentation techniques for biological sequence data. Towards this end we develop nucleotide augmentation (NTA), which leverages natural nucleotide codon degeneracy to augment protein sequence data in a biologically meaningful way. As a proof of concept for protein engineering, we apply NTA to train machine learning models with benchmark data sets of protein genotype and phenotype, revealing performance gains on par and surpassing benchmarks models, even when only using a fraction of the training data. NTA also enables substantial improvements for classification tasks under heavy class imbalance. Availability and implementationThe code to use NTA and to reproduce the analyses in this study is publicly available at https://github.com/minotm/NTA

bioinformatics↗

bioRxiv2022

Single-cell immune repertoire sequencing of B and T cells in murine models of infection and autoimmunity

Adaptive immune repertoires are composed by the ensemble of B and T cell receptors (BCR, TCR) within an individual and reflect both past and current immune responses. Recent advances in single-cell sequencing enable recovery of the complete adaptive immune receptor sequences in addition to transcriptional information. Such high-dimensional datasets enable the molecular quantification of clonal selection of B and T cells across a wide variety of conditions such as infection and disease. Due to costs, time required for the analysis and current practices of academic publishing, small-scale sequencing studies are often not made publicly available, despite having informative potential to elucidate immunological principles and guide future-studies. Here, we performed single-cell sequencing of B and T cells to profile clonal selection across murine models of viral infection and autoimmune disease. Specifically, we recovered transcriptome and immune repertoire information for polyclonal T follicular helper cells following acute and chronic viral infection, CD8+ T cells with binding specificity restricted to two distinct peptides of lymphocytic choriomeningitis virus, and B and T cells isolated from the nervous system in the context of experimental autoimmune encephalomyelitis. We could relate repertoire features such as clonal expansion, germline gene usage, and clonal convergence to cell phenotypes spanning activation, memory, naive, antibody secretion, T cell inflation, and regulation. Together, this dataset provides a resource for experimental and computational immunologists that can be integrated with future single-cell immune repertoire and transcriptome sequencing datasets.

immunology↗

bioRxiv2022

Reference-based comparison of adaptive immune receptor repertoires

B- and T-cell receptor (immune) repertoires can represent an individuals 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 (e.g., clonal diversity, germline usage). Here, we introduce immuneREF: a quantitative multi-dimensional 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. immuneREF is implemented in an R package and was validated based on detection sensitivity of immune repertoires with known similarities and dissimilarities. To quantify immune repertoire similarity landscapes across health and disease, we applied immuneREF to >2400 datasets from individuals with varying immune states (healthy, [autoimmune] disease and infection [Covid-19], immune cell population). Importantly 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 implements population-wide analysis of immune repertoire similarity and thus enables the study of the adaptive immune response across health and disease states.

systems biology↗

bioRxiv2022

A synthetic bispecific antibody capable of neutralizing SARS-CoV-2 Delta and Omicron

Bispecific antibodies have emerged as a promising strategy for curtailing severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) immune escape. This brief report highlights RBT-0813 (also known as TB493-04), a synthetic, humanized, receptor-binding domain (RBD)-targeted bispecific antibody that retains picomolar affinity to the Spike (S) trimers of all major variants of concern and neutralizes both SARS-CoV-2 Delta and Omicron in vitro.

immunology↗