Biology subjects

Robert, P. A.

Publications and source records attributed to Robert, P. A..

18 recordsLinked to original sources

bioRxiv2025

Active learning for improving out-of-distribution lab-in-the-loop experimental design

The accurate prediction of antibody-antigen binding is crucial for developing antibody-based therapeutics and advancing immunological research. Library-on-library approaches, where many antigens are probed against many antibodies, can identify specific interacting pairs. Machine learning models can predict target binding by analyzing many-to-many relationships between antibodies and antigens. However, these models face challenges when predicting interactions when test antibodies and antigens are not represented in the training data, a scenario known as out-of-distribution prediction. Generating experimental binding data is costly, limiting the availability of comprehensive datasets. Active learning can reduce costs by starting with a small labeled subset of data and iteratively expanding the labeled dataset. Few active learning approaches are available to handle data with many-to-many relationships as, for example, obtained from library-on-library screening approaches. In this study, we developed and evaluated fourteen novel active learning strategies for antibody-antigen binding prediction in a library-on-library setting and evaluated their out-of-distribution performance using the Absolut! simulation framework. We found that three of the fourteen algorithms tested significantly outperformed the baseline where random data are iteratively labeled. The best algorithm reduced the number of required antigen mutant variants by up to 35%, and sped up the learning process by 28 steps compared to the random baseline. These findings demonstrate that active learning can improve experimental efficiency in a library-on-library setting and advance antibody-antigen binding prediction.

bioinformatics↗

bioRxiv2024

Training data composition determines machine learning generalization and biological rule discovery

Supervised machine learning models depend on training datasets with positive and negative examples. Therefore, dataset composition directly impacts model performance and bias. Given the importance of machine learning for immunotherapeutic design, we examined how different negative class definitions affect model generalization and rule discovery for antibody-antigen binding. Using synthetic structure-based binding data, we evaluated models trained with various definitions of negative sets. Our findings reveal that high out-of-distribution performance can be achieved when the negative dataset contains more similar samples to the positive dataset despite a lower within-distribution performance. Furthermore, leveraging ground truth information, we show that binding rules discovered as associated with positive data change based on the negative data used. Validation on experimental data supported simulation-based observations. This work underscores the role of dataset composition, including negative data selection, in creating robust, generalizable, and biology-aware sequence-based ML models.

bioinformatics↗

bioRxiv2023

Simulation of adaptive immune receptors and repertoires with complex immune information to guide the development and benchmarking of AIRR machine learning

Machine learning (ML) has shown great potential in the adaptive immune receptor repertoire (AIRR) field. However, there is a lack of large-scale ground-truth experimental AIRR data suitable for AIRR-ML-based disease diagnostics and therapeutics discovery. Simulated ground-truth AIRR data are required to complement the development and benchmarking of robust and interpretable AIRR-ML methods where experimental data is currently inaccessible or insufficient. The challenge for simulated data to be useful is incorporating key features observed in experimental repertoires. These features, such as antigen or disease-associated immune information, cause AIRR-ML problems to be challenging. Here, we introduce LIgO, a software suite, which simulates AIRR data for the development and benchmarking of AIRR-ML methods. LIgO incorporates different types of immune information both on the receptor and the repertoire level and preserves native-like generation probability distribution. Additionally, LIgO assists users in determining the computational feasibility of their simulations. We show two examples where LIgO supports the development and validation of AIRR-ML methods: (1) how individuals carrying out-of-distribution immune information impacts receptor-level prediction performance and (2) how immune information co-occurring in the same AIRs impacts the performance of conventional receptor-level encoding and repertoire-level classification approaches. LIgO guides the advancement and assessment of interpretable AIRR-ML methods.

bioinformatics↗

bioRxiv2023

Weakly supervised identification and generation of adaptive immune receptor sequences associated with immune disease status

Adaptive immune receptor (AIR) repertoires carry immune signals as sequence motif imprints of past and present encounters with antigen (immune status). Machine learning (ML)-based identification and generation of antigen-specific immune receptors is potentially immense value for public health. The ideal training data for such ML tasks would be AIR datasets, where each sequence is labeled with its cognate antigen. However, given current technological constraints, sequence-labeled datasets are scarce, contrasted by an abundance of repertoire-labeled ones - AIR repertoire datasets where only the repertoire dataset, but not the individual AIRs, are labeled. Therefore, an unmet need exists for an ML approach that enables predictive identification and generation of disease-specific novel AIR sequences using exclusively repertoire-level immune status information. To address this need, we developed AIRRTM, an end-to-end generative model using an encoder-decoder architecture and Topic Modeling (TM) that requires exclusively repertoire-labeled AIR sequencing data as input. We validated AIRRTMs capacity to identify and generate novel disease-associated receptors on several ground truth synthetic datasets of increasingly complex immune signals and experimental data. AIRRTM broadens the discovery space for immunotherapeutics by enabling the exploitation of large-scale and broadly available immune repertoire data previously deemed largely unsuitable for this task.

immunology↗

bioRxiv2023

High-resolution mapping of cell-cycle dynamics during steady-state T-cell development and regeneration in vivo

Control of cell proliferation is critical for the lymphocyte life cycle. However, little is known on how stage-specific alterations in cell cycle behavior drive proliferation dynamics during T-cell development. Here, we employed in vivo dual-nucleoside pulse labeling combined with determination of DNA replication over time as well as fluorescent ubiquitination-based cell cycle indicator mice to establish a quantitative high-resolution map of cell cycle kinetics of thymocytes. We developed an agent-based mathematical model of T-cell developmental dynamics. To generate the capacity for proliferative bursts, cell cycle acceleration followed a stretch model, characterized by simultaneous and proportional contraction of both G1 and S phase. Analysis of cell cycle phase dynamics during regeneration showed tailored adjustments of cell cycle phase dynamics. Taken together, our results highlight intrathymic cell cycle regulation as an adjustable system to maintain physiologic tissue homeostasis and foster our understanding of dysregulation of the T-cell developmental program.

immunology↗

bioRxiv2023

Germinal centers are permissive to subdominant antibody responses

A protective humoral response to pathogens requires the development of high affinity antibodies in germinal centers (GC). The combination of antigens available during immunization has a strong impact on the strength and breadth of the antibody response. Antigens can display various levels of immunogenicity, and a hierarchy of immunodominance arises when the GC response to an antigen dampens the response to other antigens. Immunodominance is a challenge for the development of vaccines to mutating viruses, and for the development of broadly neutralizing antibodies. The extent by which antigens with different levels of immunogenicity compete for the induction of high affinity antibodies and therefore contribute to immunodominance is not known. Here, we perform in silico simulations of the GC response, using a structural representation of antigens with complex surface amino acid composition and topology. We generate antigens with different levels of immunogenicity and perform simulations with combinations of these antigens. We found that GC dynamics were driven by the most immunogenic antigen and immunodominance arose as affinity maturation to less immunogenic antigens was inhibited. However, this inhibition was moderate since the less immunogenic antigen exhibited a weak GC response in the absence of other antigens. Less immunogenic antigens reduced the dominance of GC responses to more immunogenic antigens, albeit at a later time point. The simulations suggest that increased vaccine valence may decrease immunodominance of the GC response to strongly immunogenic antigens and therefore, act as a potential strategy for the natural induction of broadly neutralizing antibodies in GC reactions.

systems biology↗

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↗

bioRxiv2021

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

Generative machine learning (ML) has been postulated to be 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 both the computation of antibody-antigen 3D-structures as well as functions as an oracle for unrestricted prospective evaluation of the antigen specificity of ML-generated antibody sequences. We found that a deep generative model, trained exclusively on antibody sequence (1D) data can be used to design native-like conformational (3D) epitope-specific antibodies, matching or exceeding the training dataset in affinity and developability variety. Furthermore, we show that transfer learning enables the generation of high-affinity antibody sequences from low-N training data. Finally, we validated that the antibody design insight gained from simulated antibody-antigen binding data is applicable to experimental real-world data. Our work establishes a priori feasibility and the theoretical foundation of high-throughput ML-based mAb design. HighlightsO_LIA large-scale dataset of 70M [3 orders of magnitude larger than the current state of the art] synthetic antibody-antigen complexes, that reflect biological complexity, allows the prospective evaluation of antibody generative deep learning C_LIO_LICombination of generative learning, synthetic antibody-antigen binding data, and prospective evaluation shows that deep learning driven antibody design and discovery at an unconstrained level is feasible C_LIO_LITransfer learning (low-N learning) coupled to generative learning shows that antibody-binding rules may be transferred across unrelated antibody-antigen complexes C_LIO_LIExperimental validation of antibody-design conclusions drawn from deep learning on synthetic antibody-antigen binding data C_LI Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=200 SRC="FIGDIR/small/451480v1_ufig1.gif" ALT="Figure 1"> View larger version (64K): org.highwire.dtl.DTLVardef@1d947bcorg.highwire.dtl.DTLVardef@1f4c51corg.highwire.dtl.DTLVardef@b8df67org.highwire.dtl.DTLVardef@1530ea5_HPS_FORMAT_FIGEXP M_FIG We leverage large synthetic ground-truth data to demonstrate the (A,B) unconstrained deep generative learning-based generation of native-like antibody sequences, (C) the prospective evaluation of conformational (3D) affinity, paratope-epitope pairs, and developability. (D) Finally, we show increased generation quality of low-N-based machine learning models via transfer learning. C_FIG

immunology↗

bioRxiv2021

A billion synthetic 3D-antibody-antigen complexes enable unconstrained machine-learning formalized investigation of antibody specificity prediction

Machine learning (ML) is a key technology for accurate prediction of antibody-antigen binding. Two orthogonal problems hinder the application of ML to antibody-specificity prediction and the benchmarking thereof: The lack of a unified ML formalization of immunological antibody specificity prediction problems and the unavailability of large-scale synthetic benchmarking datasets of real-world relevance. Here, we developed the Absolut! software suite that enables parameter-based unconstrained generation of synthetic lattice-based 3D-antibody-antigen binding structures with ground-truth access to conformational paratope, epitope, and affinity. We formalized common immunological antibody specificity prediction problems as ML tasks and confirmed that for both sequence and structure-based tasks, accuracy-based rankings of ML methods trained on experimental data hold for ML methods trained on Absolut!-generated data. The Absolut! framework thus enables real-world relevant development and benchmarking of ML strategies for biotherapeutics design. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=195 SRC="FIGDIR/small/451258v3_ufig1.gif" ALT="Figure 1"> View larger version (86K): org.highwire.dtl.DTLVardef@4c3a6aorg.highwire.dtl.DTLVardef@8f1deborg.highwire.dtl.DTLVardef@1f55ec6org.highwire.dtl.DTLVardef@2135dd_HPS_FORMAT_FIGEXP M_FIG The software framework Absolut! enables (A,B) the generation of virtually arbitrarily large numbers of synthetic 3D-antibody-antigen structures, (C,D) the formalization of antibody specificity as machine learning (ML) tasks as well as the exploration of ML strategies for real-world antibody-antigen binding or paratope-epitope prediction. C_FIG HighlightsO_LISoftware framework Absolut! to generate an arbitrarily large number of synthetic 3D-antibody-antigen structures that contain biological layers of antibody-antigen binding complexity that render ML predictions challenging C_LIO_LIImmunological antibody specificity prediction problems formalized as machine learning tasks for which the in silico complexes are immediately usable as benchmark datasets C_LIO_LIExploration of machine learning prediction accuracy as a function of architecture, dataset size, choice of negatives, and sequence-structure encoding C_LIO_LIRelative ML performance learnt on Absolut! datasets transfers to experimental datasets C_LI

immunology↗

bioRxiv2021

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 individuals 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 non-genetic 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.

immunology↗

bioRxiv2021

immuneML: an ecosystem for machine learning analysis of adaptive immune receptor repertoires

Adaptive immune receptor repertoires (AIRR) are key targets for biomedical research as they record past and ongoing adaptive immune responses. The capacity of machine learning (ML) to identify complex discriminative sequence patterns renders it an ideal approach for AIRR-based diagnostic and therapeutic discovery. To date, widespread adoption of AIRR ML has been inhibited by a lack of reproducibility, transparency, and interoperability. immuneML (immuneml.uio.no) addresses these concerns by implementing each step of the AIRR ML process in an extensible, open-source software ecosystem that is based on fully specified and shareable workflows. To facilitate widespread user adoption, immuneML is available as a command-line tool and through an intuitive Galaxy web interface, and extensive documentation of workflows is provided. We demonstrate the broad applicability of immuneML by (i) reproducing a large-scale study on immune state prediction, (ii) developing, integrating, and applying a novel method for antigen specificity prediction, and (iii) showcasing streamlined interpretability-focused benchmarking of AIRR ML.

bioinformatics↗

bioRxiv2020

Influenza A virus-induced thymus atrophy differentially affects dynamics of conventional and regulatory T cell development

Foxp3+regulatory T (Treg) cells, which are crucial for maintenance of self-tolerance, mainly develop within the thymus, where they arise from CD25+Foxp3-or CD25-Foxp3+ Treg cell precursors. Although it is known that infections can cause transient thymic involution, the impact of infection-induced thymus atrophy on thymic Treg (tTreg) cell development is unknown. Here, we infected mice with influenza A virus (IAV) and studied thymocyte population dynamics post infection. IAV infection caused a massive, but transient thymic involution, dominated by a loss of CD4+CD8+ double-positive (DP) thymocytes, which was accompanied by a significant increase in the frequency of CD25+Foxp3+ tTreg cells. Differential apoptosis susceptibility could be experimentally excluded as a reason for the relative tTreg cell increase, and mathematical modeling suggested that enhanced tTreg cell generation cannot explain the increased frequency of tTreg cells. Yet, an increased death of DP thymocytes and augmented exit of single-positive (SP) thymocytes was suggested to be causative. Interestingly, IAV-induced thymus atrophy resulted in a significantly reduced T cell receptor (TCR) repertoire diversity of newly produced tTreg cells. Taken together, IAV-induced thymus atrophy is substantially altering the dynamics of major thymocyte populations, finally resulting in a relative increase of tTreg cells with an altered TCR repertoire.

immunology↗

bioRxiv2020

Agent-Based Modeling of T Cell Receptor Cooperativity

Immunological synapse (IS) formation is a key event during antigen recognition by T cells. Recent experimental evidence suggests that the affinity between T cell receptors (TCRs) and antigen is actively modulated during the early steps of TCR signaling. In this work, we used an agent-based model to study possible mechanisms for affinity modulation during IS formation. We show that, without any specific active mechanism, the observed affinity between receptors and ligands evolves over time, and depends on the density of ligand pMHC (antigen peptide presented by major histocompatibility complexes) and TCR molecules. Comparison between the presence or absence of TCR-pMHC centrally directed flow due to F-actin coupling suggest centripetal transport is a potential mechanism for the affinity modulation. The model further suggests that the time point of affinity measurement during immune synapse formation is critical. Finally, a mathematical model of F-actin foci formation incorporated in the agent-based model, shows that TCR affinity can potentially be actively modulated by a positive/negative feedback of F-actin foci on the TCR-pMHC association rate kon.Competing Interest StatementThe authors have declared no competing interest.AbbreviationsView Full Text

systems biology↗

bioRxiv2020

In silico characterization of mechanisms positioning costimulatory and checkpoint complexes in immune synapses

Integrin and small immunoglobulin superfamily (sIGSF) adhesion complexes function physiologically in human immunological synapses (IS) wherein sIGSF complexes form a corolla of microdomains around an integrin ring and secretory core. The corolla recruits and retains the major costimulatory and checkpoint complexes that regulate the response to T cell receptor (TCR) engagement, making forces that govern corolla formation of particular interest. We developed a phenomenological agent-based model in order to test different hypotheses concerning the mechanisms underlying molecular reorganization during IS formation. The model showed that sIGSF complexes are passively excluded to the distal aspect of the IS as long as their interaction with the ramified F-actin transport network is absent or weaker than that of integrins. An attractive force between sIGSF adhesion and costimulatory/checkpoint complexes relocates the latter from the centre of the IS to the corolla. The simulations suggest that size based sorting interactions with large glycocalyx components as well as a short-range self-attraction between sIGSF complexes explain the corolla "petals". These molecular and mechanistic features establish a general model that can recapitulate complex pattern formation processes observed in cell-bilayer and cell-cell interfaces. One Sentence SummaryComputer simulations of immunological synapses reveal the localization mechanisms of immunoglobulin superfamily adhesion and costimulatory/checkpoint complexes.

immunology↗

bioRxiv2019

Evaluation of CD8 T cell killing models with computer simulations of 2-photon imaging experiments

In vivo imaging of cytotoxic T lymphocyte (CTL) killing activity revealed that infected cells have a higher observed probability of dying after multiple contacts with CTLs, suggesting memory effect in CTLs or infected cells. We developed a three-dimensional agent-based model of CTL killing activity to discriminate different hypotheses about how infected cells get killed based on quantitative 2-photon in vivo observations. We compared a constant CTL killing probability with mechanisms of signal integration in CTL or infected cells. The most likely scenario implied increased susceptibility of infected cells with increasing number of CTL contacts where the total number of contacts was a critical factor as opposed to signal integration over many contacts. However, when allowing in silico T cells to interact with apoptotic target cells (zombie contacts), a contact history independent killing mechanism was also in agreement with the experimental datasets. We showed that contacts that take place between CTLs and dying infected cells impact the observed killing dynamics because even in absence of modulation of cell properties, we saw an increase of the observed probability of killing infected cells with more interactions. The duration taken by an infected cell to die and the per capita killing rate (PCKR) of CTLs, parameters hard to measure directly, were determined from the model and turned out predictive to distinguish the different CTL killing models in future experiments. The comparison of observed datasets to simulation results, revealed limitations in interpreting 2-photon data, and provided prediction for additional measurements to distinguish CTL killing models. HighlightsO_LIKilling of infected cells by cytotoxic T cells typically involves more than a single contact. C_LIO_LICytotoxic T cells or infected cells integrate signals from multiple interactions. C_LIO_LIT cell contacts with dying infected cells have a major impact on in vivo data interpretation. C_LI Significance StatementDespite having a clear understanding of cytotoxic T lymphocyte (CTL) mediated cytotoxicity mechanisms, the quantitative dynamics remain unexplored at a cellular level. We developed an agent-based model to compare different hypotheses for mechanisms of CTL mediated cytotoxicity that could lead to an increase in observed probability of killing infected cells at higher interactions with CTLs as seen in vivo. We showed that this behaviour can be explained by modulation of properties by infected cells or CTLs with increasing number of contacts. For the modulation, we compared two modes of signal integration and showed that time is not a relevant parameter in signal integration. We also studied the impact of contacts between CTLs and apoptotic infected cells on observed killing properties.

systems biology↗

bioRxiv2019

A 3D structural affinity model for multi-epitope in silico germinal center simulations

Vaccine development for mutating pathogens is challenged by their fast evolution, the complexity of immunodominance, and the heterogeneous immune history of individuals. Mathematical models are critical for predicting successful vaccine conditions or designing potent antibodies. Existing models are limited by their abstract and poorly structural representations of antigen epitopes. Here, we propose a structural lattice-based model for antibody-antigen affinity. An efficient algorithm is given that predicts the best binding structure of an antibodys amino acid sequence around an antigen with shortened computational time. This structural representation contains key physiological properties, such as affinity jumps and cross-reactivity, and successfully reflects the topology of antigen epitopes, such as pockets and shielded residues. It is suitable for large simulations of affinity maturation. We perform in silico immunizations via germinal center simulations and show that our model can explain complex phenomena like recognition of the same epitope by unrelated clones. We show that the use of cocktails of similar epitopes promotes the development of cross-reactive antibodies. This model opens a new avenue for optimizing multivalent vaccines with combined antigen cocktails or sequential immunizations, and to reveal reasons for vaccine success or failure on a structural basis.

systems biology↗

bioRxiv2019

A finite vocabulary of antibody-antigen interaction enables predictability of paratope-epitope binding

Antibody-antigen binding relies on the specific interaction of amino acids at the paratope-epitope interface. The predictability of antibody-antigen binding is a prerequisite for de novo antibody and (neo-)epitope design. A fundamental premise for the predictability of antibody-antigen binding is the existence of paratope-epitope interaction motifs that are universally shared among antibody-antigen structures. In the largest set of non-redundant antibody-antigen structures, we identified structural interaction motifs, which together compose a commonly shared structure-based vocabulary of paratope-epitope interactions. We show that this vocabulary enables the machine learnability of antibody-antigen binding on the paratope-epitope level using generative machine learning. The vocabulary (i) is compact, less than 104 motifs, (ii) distinct from non-immune protein-protein interactions, and (iii) mediates specific oligo- and polyreactive interactions between paratope-epitope pairs. Our work successfully leveraged combined structure- and sequence-based learning showing that machine-learning-driven predictive paratope and epitope engineering is feasible.

immunology↗

bioRxiv2017

Quantitative characterization of CTLA4 trafficking and turnover using a combined in vitro and in silico approach

CTLA4 is an essential negative regulator of T cell immune responses and is a key checkpoint regulating autoimmunity and anti-tumour immunity. Genetic mutations resulting in a quantitative defect in CTLA4 are associated with the development of an immune dysregulation syndrome. Endocytosis of CTLA4 is rapid and continuous with subsequent degradation or recycling. CTLA4 has two natural ligands, the surface transmembrane proteins CD80 and CD86 that are shared with the T cell co-stimulatory receptor CD28. Upon ligation with CD80/CD86, CTLA4 can remove these ligands from the opposing cells by transendocytosis. The efficiency of ligand removal is thought to be highly dependent on the processes involved in CTLA4 trafficking. With a combined in vitro-in silico study, we quantify the rates of CTLA4 internalization, recycling and degradation. We incorporate experimental data from cell lines and primary human T cells. Our model provides a framework for exploring the impact of altered affinity of natural ligands or therapeutic anti-CTLA4 antibodies and for predicting the effect of clinically relevant CTLA4 pathway mutations. The presented methodology for extracting trafficking rates can be transferred to the study of other transmembrane proteins.

immunology↗