Linguistics-based formalization of the antibody language as a basis for antibody language models.

Apparent parallels between natural language and antibody sequences have led to a surge in deep language models applied to antibody sequences for predicting cognate antigen recognition. However, a linguistic formal definition of antibody language does not exist, and insight into how antibody language models capture antibody-specific binding features remains largely uninterpretable. Here we describe how a linguistic formalization of the antibody language, by characterizing its tokens and grammar, could address current challenges in antibody language model rule mining.

Corrigendum: Experimental validation of immunogenic SARS-CoV-2 T cell epitopes identified by artificial intelligence.

[This corrects the article DOI: 10.3389/fimmu.2023.1265044.].

Experimental validation of immunogenic SARS-CoV-2 T cell epitopes identified by artificial intelligence.

During the COVID-19 pandemic we utilized an AI-driven T cell epitope prediction tool, the NEC Immune Profiler (NIP) to scrutinize and predict regions of T cell immunogenicity (hotspots) from the entire SARS-CoV-2 viral proteome. These immunogenic regions offer potential for the development of universally protective T cell vaccine candidates. Here, we validated and characterized T cell responses to a set of minimal epitopes from these AI-identified universal hotspots. Utilizing a flow cytometry-based T cell activation-induced marker (AIM) assay, we identified 59 validated screening hits, of which 56% (33 peptides) have not been previously reported. Notably, we found that most of these novel epitopes were derived from the non-spike regions of SARS-CoV-2 (Orf1ab, Orf3a, and E). In addition, ex vivo stimulation with NIP-predicted peptides from the spike protein elicited CD8+ T cell response in PBMC isolated from most vaccinated donors. Our data confirm the predictive accuracy of AI platforms modelling bona fide immunogenicity and provide a novel framework for the evaluation of vaccine-induced T cell responses.

Toward real-world automated antibody design with combinatorial Bayesian optimization.

Antibodies are multimeric proteins capable of highly specific molecular recognition. The complementarity determining region 3 of the antibody variable heavy chain (CDRH3) often dominates antigen-binding specificity. Hence, it is a priority to design optimal antigen-specific CDRH3 to develop therapeutic antibodies. The combinatorial structure of CDRH3 sequences makes it impossible to query binding-affinity oracles exhaustively. Moreover, antibodies are expected to have high target specificity and developability. Here, we present AntBO, a combinatorial Bayesian optimization framework utilizing a CDRH3 trust region for an in silico design of antibodies with favorable developability scores. The in silico experiments on 159 antigens demonstrate that AntBO is a step toward practically viable in vitro antibody design. In under 200 calls to the oracle, AntBO suggests antibodies outperforming the best binding sequence from 6.9 million experimentally obtained CDRH3s. Additionally, AntBO finds very-high-affinity CDRH3 in only 38 protein designs while requiring no domain knowledge.

Machine learning and molecular simulation ascertain antimicrobial peptide against Klebsiella pneumoniae from public database.

Antimicrobial peptides (AMPs) are short peptides with a broad spectrum of antimicrobial activity. They play a key role in the host innate immunity of many organisms. The growing threat of microorganisms resistant to antimicrobial agents and the lack of new commercially available antibiotics have made in silico discovery of AMPs increasingly important. Machine learning (ML) has improved the speed and efficiency of AMP discovery while reducing the cost of experimental approaches. Despite various ML platforms developed, there is still a lack of integrative use of ML platforms for AMP discovery from publicly available protein databases. Therefore, our study aims to screen potential AMPs with antibiofilm properties from databases using ML platforms, followed by protein-peptide molecular docking analysis and molecular dynamics (MD) simulations. A total of 5850 peptides classified as non-AMP were screened from UniProtKB and analyzed using various online ML platforms (e.g., CAMPr3, DBAASP, dPABBs, Hemopred, and ToxinPred). Eight potential AMP peptides against Klebsiella pneumoniae with antibiofilm, non-toxic and non-hemolytic properties were then docked to MrkH, a transcriptional regulator of type 3 fimbriae involved in biofilm formation. Five of eight peptides bound more strongly than the native MrkH ligand when analyzed using HADDOCK and HPEPDOCK. Following the docking studies, our MD simulated that a Neuropeptide B (Peptide 3) bind strongly to the MrkH active sites. The discovery of putative AMPs that exceed the binding energies of the native ligand underscores the utility of the combined ML and molecular simulation strategies for discovering novel AMPs with antibiofilm properties.

Reference-based comparison of adaptive immune receptor repertoires.

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

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

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

Progress and challenges for the machine learning-based design of fit-for-purpose monoclonal antibodies.

Although the therapeutic efficacy and commercial success of monoclonal antibodies (mAbs) are tremendous, the design and discovery of new candidates remain a time and cost-intensive endeavor. In this regard, progress in the generation of data describing antigen binding and developability, computational methodology, and artificial intelligence may pave the way for a new era of in silico on-demand immunotherapeutics design and discovery. Here, we argue that the main necessary machine learning (ML) components for an in silico mAb sequence generator are: understanding of the rules of mAb-antigen binding, capacity to modularly combine mAb design parameters, and algorithms for unconstrained parameter-driven in silico mAb sequence synthesis. We review the current progress toward the realization of these necessary components and discuss the challenges that must be overcome to allow the on-demand ML-based discovery and design of fit-for-purpose mAb therapeutic candidates.

Unconstrained generation of synthetic antibody-antigen structures to guide machine learning methodology for 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 datasets to benchmark real-world relevant ML methods and dataset design. Here we developed the Absolut! software suite that enables parameter-based unconstrained generation of synthetic lattice-based three-dimensional 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 has the potential to enable real-world relevant development and benchmarking of ML strategies for biotherapeutics design.

Individualized VDJ recombination predisposes the available Ig sequence space.

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

A compact vocabulary of paratope-epitope interactions enables predictability of antibody-antigen 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 a dataset of non-redundant antibody-antigen structures, we identify 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 (1) is compact, less than 104 motifs; (2) distinct from non-immune protein-protein interactions; and (3) mediates specific oligo- and polyreactive interactions between paratope-epitope pairs. Our work leverages combined structure- and sequence-based learning to demonstrate that machine-learning-driven predictive paratope and epitope engineering is feasible.

The immuneML 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 deep learning method for antigen specificity prediction, and (iii) showcasing streamlined interpretability-focused benchmarking of AIRR ML.

Interorganelle Tethering to Endocytic Organelles Determines Directional Cytokine Transport in CD4+ T Cells.

Delivery of vesicles to their desired destinations plays a central role in maintaining proper cell functionality. In certain scenarios, depending on loaded cargos, the vesicles have spatially distinct destinations. For example, in T cells, some cytokines (e.g., IL-2) are polarized to the T cell-target cell interface, whereas the other cytokines are delivered multidirectionally (e.g., TNF-α). In this study, we show that in primary human CD4+ T cells, both TNF-α+ and IL-2+ vesicles can tether with endocytic organelles (lysosomes/late endosomes) by forming membrane contact sites. Tethered cytokine-containing vesicle (CytV)-endocytic organelle pairs are released sequentially. Only endocytic organelle-tethered CytVs are preferentially transported to their desired destination. Mathematical models suggest that endocytic organelle tethering could regulate the direction of cytokine transport by selectively attaching different microtubule motor proteins (such as kinesin and dynein) to the corresponding CytVs. These findings establish the previously unknown interorganelle tethering to endocytic organelles as a universal solution for directional cytokine transport in CD4+ T cells. Modulating tethering to endocytic organelles can, therefore, coordinately control directionally distinct cytokine transport.

immuneSIM: tunable multi-feature simulation of B- and T-cell receptor repertoires for immunoinformatics benchmarking.

SUMMARY: B- and T-cell receptor repertoires of the adaptive immune system have become a key target for diagnostics and therapeutics research. Consequently, there is a rapidly growing number of bioinformatics tools for immune repertoire analysis. Benchmarking of such tools is crucial for ensuring reproducible and generalizable computational analyses. Currently, however, it remains challenging to create standardized ground truth immune receptor repertoires for immunoinformatics tool benchmarking. Therefore, we developed immuneSIM, an R package that allows the simulation of native-like and aberrant synthetic full-length variable region immune receptor sequences by tuning the following immune receptor features: (i) species and chain type (BCR, TCR, single and paired), (ii) germline gene usage, (iii) occurrence of insertions and deletions, (iv) clonal abundance, (v) somatic hypermutation and (vi) sequence motifs. Each simulated sequence is annotated by the complete set of simulation events that contributed to its in silico generation. immuneSIM permits the benchmarking of key computational tools for immune receptor analysis, such as germline gene annotation, diversity and overlap estimation, sequence similarity, network architecture, clustering analysis and machine learning methods for motif detection. AVAILABILITY AND IMPLEMENTATION: The package is available via https://github.com/GreiffLab/immuneSIM and on CRAN at https://cran.r-project.org/web/packages/immuneSIM. The documentation is hosted at https://immuneSIM.readthedocs.io. CONTACT: sai.reddy@ethz.ch or victor.greiff@medisin.uio.no. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

ALLO: A tool to discriminate and prioritize allosteric pockets.

Allosteric proteins make up a substantial proportion of human drug targets. Thus, rational design of small molecule binders that target these proteins requires the identification of putative allosteric pockets and an understanding of their potential activity. Here, we characterized allosteric pockets using a set of physicochemical descriptors and compared them to pockets that are found on the surface of a protein. Further, we trained predictive models capable of discriminating allosteric pockets from orthosteric pockets and models capable of prioritizing allosteric pockets in a set of pockets found on a given protein. Such models might be useful for identifying novel allosteric sites and in turn, potentially new allosteric drug targets. Datasets along with a Python program encapsulating the predictive models are available at http://github.com/fibonaccirabbits/allo.

ENRI: A tool for selecting structure-based virtual screening target conformations.

Finding pharmaceutically relevant target conformations from an arbitrary set of protein conformations remains a challenge in structure-based virtual screening (SBVS). The growth in the number of available conformations, either experimentally determined or computationally derived, obscures the situation further. While the inflated conformation space potentially contains viable druggable targets, the increase of conformational complexity, as a consequence, poses a selection problem. To address this challenge, we took advantage of machine learning methods, namely an over-sampling and a binary classification procedure, and present a novel method to select druggable receptor conformations. Specifically, we trained a binary classifier on a set of nuclear receptor conformations, wherein each conformation was labeled with an enrichment measure for a corresponding SBVS. The classifier enabled us to formulate suggestions and identify enriching SBVS targets for six of seven nuclear receptors. Further, the classifier can be extended to other proteins of interest simply by feeding new training data sets to the classifier. Our work, thus, provides a methodology to identify pharmaceutically interesting receptor conformations for nuclear receptors and other drug targets.

Stability, orientation and position preference of the stem region (residues 689-703) in Hepatitis C Virus (HCV) envelope glycoprotein E2: a molecular dynamics study.

Envelope glycoproteins of Hepatitis C Virus (HCV) play an important role in the virus assembly and initial entry into host cells. Conserved charged residues of the E2 transmembrane (TM) domain were shown to be responsible for the heterodimerization with envelope glycoprotein E1. Despite intensive research on both envelope glycoproteins, the structural information is still not fully understood. Recent findings have revealed that the stem (ST) region of E2 also functions in the initial stage of the viral life cycle. We have previously shown the effect of the conserved charged residues on the TM helix monomer of E2. Here, we extended the model of the TM domain by adding the adjacent ST segment. Explicit molecular dynamics simulations were performed for the E2 amphiphilic segment of the ST region connected to the putative TM domain (residues 683-746). Structural conformation and behavior are studied and compared with the nuclear magnetic resonance (NMR)-derived segment of E2 ( 2KQZ.pdb). We observed that the central helix of the ST region (residues 689 - 703) remained stable as a helix in-plane to the lipid bilayer. Furthermore, the TM domain appeared to provide minimal contribution to the structural stability of the amphipathic region. This study also provides insight into the orientation and positional preferences of the ST segment with respect to the membrane lipid-water interface.

Interaction of ganoderic acid on HIV related target: molecular docking studies.

Finding the ultimate HIV cure remain a challenging tasks for decades. Various active compounds have been tested against various components of the virus in the effort to halt the virus development in infected host. The idea of finding cure from known pharmacologically active natural occurring compounds is intriguing and practical. Ganoderma lucidum (Ling-Zhi or Reishi) is one of the most productive and pharmacologically active compounds found in Asian countries. It has been used traditionally for many years throughout different cultures. More than a decade ago, el-Mekkawy and co-workers (1998) have tested several active compounds found in this plant. They have successfully identified several active compounds with reasonable inhibitory activity against HIV protease however; no further studies were done on these compounds. This study aimed to elucidate interactions for one of the active compounds of Ganoderma lucidum namely ganoderic acid with HIV-1 protease using molecular docking simulation. This study revealed four hydrogen bonds formed between model34 of ganoderic acid B and 1HVR. Hydrogen bonds in 1HVR-Model34 complex were formed through ILE50, ILE50', ASP29 and ASP30 residues. Interestingly similar interactions were also observed in the native ligand in 1HVR. Furthermore, interactions involving ILE50 and ILE50' residues have been previously identified to play central roles in HIV-1 protease-ligand interactions.These observed interactions not only suggested HIV-1 protease in general is a suitable target for ganoderic acid B, they also indicated a huge potential for HIV drug discovery based on this compound.