Accurate modeling of peptide-MHC structures with AlphaFold

Promoting lignocellulosic biorefinery by machine learning: progress, perspectives and challenges

The lignocellulosic biorefinery involves pretreatment, enzymatic hydrolysis, mixed sugar fermentation, and optional anaerobic digestion. This pipeline could be effectively implemented through machine learning (ML)-guided process optimization and strain modification rather than experimental or experience-based ones. This review takes a holistic perspective on the entire pipeline, discussing how ML could aid lignocellulosic, while other published work has focused on individual modules within the pipeline. This review also explores the model construction and evaluation strategies and highlights the emerging potential of transfer learning and hybrid ML models to address data insufficiency and improve model interpretability. Furthermore, challenges and future prospects of ML in lignocellulosic biorefinery will be elaborated in this review. Integrating ML into lignocellulosic biorefinery offers a promising pathway towards sustainable and competitive biorefinery systems.

High Throughput Meta-analysis of Antimicrobial Peptides for Characterizing Class Specific Therapeutic Candidates: An In Silico Approach

The increasing incidence of antimicrobial resistance is becoming a serious concern worldwide and requires newer drugs. Recent evidence has shown growing interest in peptide-based therapeutics. Here, we performed a meta-analysis of nearly 867,000 predicted antimicrobial peptides and assessed their antibacterial (ABPs), antifungal (AFPs), and antiviral (AVPs) activity. We created high-quality, class-specific datasets and performed several computational analyses. Composition analysis revealed enrichment of aliphatic (V, A, I, and L) and positively charged (K and R) amino acids in ABPs: aliphatic (G, I), basic (K and R), and aromatic amino acids (F) in AFPs and sulfur containing (M) and aliphatic amino acids (V, I, and L) in AVPs. We observed significant differences in the molecular weight, charge, isoelectric point, and instability index of the peptides among three classes. We observed AFPs possessing the highest molecular weight and ABPs showing the highest charge and isoelectric point, whereas instability index was found to be comparable among the three classes. Motif analysis shows enrichment of unique motifs such as "VRVR" and "AKKPA" in ABPs, "DFFAI" and "FFAI" in AFPs, and "VVV" and "IM" in AVPs. We further developed seven distinct machine learning models to predict peptide activity where ExtraTree model achieved the highest AUROC of 0.98 in classifying ABPs and non-ABPs, 0.99 for classifying AFPs and non-AFPs, and 0.99 for classifying AVPs and non-AVPs on an independent dataset. To assist scientific community, we have provided the dataset and models at our GitHub page ( https://github.com/agrawalpiyush-srm/AMP_MetaAnalysis ). Subsequent filtering of peptides based on moonlighting properties (toxicity, allergenicity, cell-penetrating ability, half-life, and secondary structure) yielded a list of peptides that exhibit substantial therapeutic potential. We further selected the top ten peptides in each class, predicted their 3D structures using ColabFold embedded in ChimeraX1.8 software and performed molecular docking analysis with a pathogenic protein selected from an organism in each class using HDOCK webserver. Docking studies demonstrated strong interaction between peptides and the proteins. Lastly, we proposed list of peptides with high therapeutic potential in each class.

Unveiling the influence of fastest nobel prize winner discovery: alphafold's algorithmic intelligence in medical sciences

CONTEXT

AlphaFold's advanced AI technology has transformed protein structure interpretation. By predicting three-dimensional protein structures from amino acid sequences, AlphaFold has solved the complex protein-folding problem, previously challenging for experimental methods due to numerous possible conformations. Since its inception, AlphaFold has introduced several versions, including AlphaFold2, AlphaFold DB, AlphaFold Multimer, Alpha Missense, and AlphaFold3, each further enhancing protein structure prediction. Remarkably, AlphaFold is recognized as the fastest Nobel Prize winner in science history. This technology has extensive applications, potentially transforming treatment and diagnosis in medical sciences by reducing drug design costs and time, while elucidating structural pathways of human body systems. Numerous studies have demonstrated how AlphaFold aids in understanding health conditions by providing critical information about protein mutations, abnormal protein-protein interactions, and changes in protein dynamics. Researchers have also developed new technologies and pipelines using different versions of AlphaFold to amplify its potential. However, addressing existing limitations is crucial to maximizing AlphaFold's capacity to redefine medical research. This article reviews AlphaFold's impact on five key aspects of medical sciences: protein mutation, protein-protein interaction, molecular dynamics, drug design, and immunotherapy.

METHODS

This review examines the contributions of various AlphaFold versions AlphaFold2, AlphaFold DB, AlphaFold Multimer, Alpha Missense, and AlphaFold3 to protein structure prediction. The methods include an extensive analysis of computational techniques and software used in interpreting and predicting protein structures, emphasizing advances in AI technology and its applications in medical research.

Neoantigen-based immunotherapy: advancing precision medicine in cancer and glioblastoma treatment through discovery and innovation

Neoantigen-based immunotherapy has emerged as a transformative approach in cancer treatment, offering precision medicine strategies that target tumor-specific antigens derived from genetic, transcriptomic, and proteomic alterations unique to cancer cells. These neoantigens serve as highly specific targets for personalized therapies, promising more effective and tailored treatments. The aim of this article is to explore the advances in neoantigen-based therapies, highlighting successful treatments such as vaccines, tumor-infiltrating lymphocyte (TIL) therapy, T-cell receptor-engineered T cells therapy (TCR-T), and chimeric antigen receptor T cells therapy (CAR-T), particularly in cancer types like glioblastoma (GBM). Advances in technologies such as next-generation sequencing, RNA-based platforms, and CRISPR gene editing have accelerated the identification and validation of neoantigens, moving them closer to clinical application. Despite promising results, challenges such as tumor heterogeneity, immune evasion, and resistance mechanisms persist. The integration of AI-driven tools and multi-omic data has refined neoantigen discovery, while combination therapies are being developed to address issues like immune suppression and scalability. Additionally, the article discusses the ongoing development of personalized immunotherapies targeting tumor mutations, emphasizing the need for continued collaboration between computational and experimental approaches. Ultimately, the integration of cutting-edge technologies in neoantigen research holds the potential to revolutionize cancer care, offering hope for more effective and targeted treatments.

A frameshift-generated cancer neoepitope that controls tumor burden in prophylaxis as well as therapy

Insertion or deletion of one or two base pairs within a coding region causes a frameshift, which has the potential to generate neoepitopes (InDel-generated neoepitopes) that lack a self-counterpart and are entirely novel. Despite the obvious appeal of InDel-generated neoepitopes, and the demonstration of such candidate neoepitopes that can elicit a CD8 T-cell response, no InDel-generated neoepitopes that actually control tumors in vivo have been reported thus far. Here, in a mouse colon carcinoma line, we identify 11 InDels, only one of which generates a neoepitope that elicits tumor control in vivo in models of prophylaxis as well as therapy. Although this neoepitope has no self-counterpart, it has a low affinity (IC50 33,937.60 nM) for its MHC I allele. Despite its low affinity for MHC I, this neoepitope elicits antitumor activity in vivo through CD8 T cells. Furthermore, CD8 T cells elicited by this InDel-generated neoepitope, like the neoepitopes created by point mutations, show notably less exhaustion than classical immunogenic epitopes. Ironically, this InDel-generated neoepitope follows the same rules as noted for most of the tumor control-mediating neoepitopes generated by point mutations that have a poor affinity for MHC I alleles.

Molecular Modelling in Bioactive Peptide Discovery and Characterisation

Molecular modelling is a vital tool in the discovery and characterisation of bioactive peptides, providing insights into their structural properties and interactions with biological targets. Many models predicting bioactive peptide function or structure rely on their intrinsic properties, including the influence of amino acid composition, sequence, and chain length, which impact stability, folding, aggregation, and target interaction. Homology modelling predicts peptide structures based on known templates. Peptide-protein interactions can be explored using molecular docking techniques, but there are challenges related to the inherent flexibility of peptides, which can be addressed by more computationally intensive approaches that consider their movement over time, called molecular dynamics (MD). Virtual screening of many peptides, usually against a single target, enables rapid identification of potential bioactive peptides from large libraries, typically using docking approaches. The integration of artificial intelligence (AI) has transformed peptide discovery by leveraging large amounts of data. AlphaFold is a general protein structure prediction tool based on deep learning that has greatly improved the predictions of peptide conformations and interactions, in addition to providing estimates of model accuracy at each residue which greatly guide interpretation. Peptide function and structure prediction are being further enhanced using Protein Language Models (PLMs), which are large deep-learning-derived statistical models that learn computer representations useful to identify fundamental patterns of proteins. Recent methodological developments are discussed in the context of canonical peptides, as well as those with modifications and cyclisations. In designing potential peptide therapeutics, the main outstanding challenge for these methods is the incorporation of diverse non-canonical amino acids and cyclisations.

Predicting reverse-bound peptide conformations in MHC Class II with PANDORA

Recent discoveries have transformed our understanding of peptide binding in Major Histocompatibility Complex (MHC) molecules, showing that peptides, for some MHC class II alleles, can bind in a reverse orientation (C-terminus to N-terminus) and can still effectively activate CD4+ T cells. These finding challenges established concepts of immune recognition and suggests new pathways for therapeutic intervention, such as vaccine design. We present an updated version of PANDORA, which, to the best of our knowledge, is the first tool capable of modeling reversed-bound peptides. Modeling these peptides presents a unique challenge due to the limited structural data available for these orientations in existing databases. PANDORA has overcome this challenge through integrative modeling using algorithmically reversed peptides as templates. We have validated the new PANDORA feature through two targeted experiments, achieving an average backbone binding-core L-RMSD value of 0.63 Å. Notably, it maintained low RMSD values even when using templates from different alleles and peptide sequences. Our results suggest that PANDORA will be an invaluable resource for the immunology community, aiding in the development of targeted immunotherapies and vaccine design.

An in-depth review of AI-powered advancements in cancer drug discovery

The convergence of artificial intelligence (AI) and genomics is redefining cancer drug discovery by facilitating the development of personalized and effective therapies. This review examines the transformative role of AI technologies, including deep learning and advanced data analytics, in accelerating key stages of the drug discovery process: target identification, drug design, clinical trial optimization, and drug response prediction. Cutting-edge tools such as DrugnomeAI and PandaOmics have made substantial contributions to therapeutic target identification, while AI's predictive capabilities are driving personalized treatment strategies. Additionally, advancements like AlphaFold highlight AI's capacity to address intricate challenges in drug development. However, the field faces significant challenges, including the management of large-scale genomic datasets and ethical concerns surrounding AI deployment in healthcare. This review underscores the promise of data-centric AI approaches and emphasizes the necessity of continued innovation and interdisciplinary collaboration. Together, AI and genomics are charting a path toward more precise, efficient, and transformative cancer therapeutics.

Cancer Immunogenomics Approaches and Applications to Cancer Vaccines

The application of next-generation sequencing-based genomics and corresponding analytical pipelines have significantly improved our ability to identify tumor-unique antigenic peptides ("neoantigens") for the design of personalized vaccine therapies and to monitor immune responses to these vaccines. The more recent implementation of artificial intelligence and machine learning into several of the more complex analytical components of the neoantigen selection process has provided significant improvements across a number of previously difficult aspects within neoantigen identification, as we will describe. Related technologies and analytics have been developed that enable the characterization of changes to the tumor immune microenvironment facilitated by vaccination and monitor systemic responses in patients. Here, we review these new methods and their application to the design, implementation, and evaluation of cancer vaccines.

AlphaFold and what is next: bridging functional, systems and structural biology

INTRODUCTION

The DeepMind's AlphaFold (AF) has revolutionized biomedical and biocience research by providing both experts and non-experts with an invaluable tool for predicting protein structures. However, while AF is highly effective for predicting structures of rigid and globular proteins, it is not able to fully capture the dynamics, conformational variability, and interactions of proteins with ligands and other biomacromolecules.

AREAS COVERED

In this review, we present a comprehensive overview of the latest advancements in 3D model predictions for biomacromolecules using AF. We also provide a detailed analysis its of strengths and limitations, and explore more recent iterations, modifications, and practical applications of this strategy. Moreover, we map the path forward for expanding the landscape of AF toward predicting structures of every protein and peptide, and their interactions in the proteome in the most physiologically relevant form. This discussion is based on an extensive literature search performed using PubMed and Google Scholar.

EXPERT OPINION

While significant progress has been made to enhance AF's modeling capabilities, we argue that a combined approach integrating both various in silico and in vitro methods will be most beneficial for the future of structural biology, bridging the gaps between static and dynamic features of proteins and their functions.

Molecular docking to investigate HLA-associated idiosyncratic drug reactions

Idiosyncratic drug reactions (IDRs) pose severe threats to patient health. Unlike conventionally dose-dependent side effects, they are unpredictable and more frequently manifest as life-threatening conditions, such as severe cutaneous adverse reactions (SCARs) and drug-induced liver injury (DILI). Some HLA alleles, such as HLA-B*57:01, HLA-B*15:02, and HLA-B*58:01, are known risk factors for adverse reactions induced by multiple drugs. However, the structural basis underlying most HLA-associated adverse events remains poorly understood. This review summarizes the application of molecular docking to reveal the mechanisms of IDR-related HLA associations, covering studies using this technique to examine drug-HLA binding pockets and identify key binding residues. We provide a comprehensive overview of risk HLA alleles associated with IDRs, followed by a discussion of the utility and limitations of commonly used molecular docking tools in simulating complex molecular interactions within the HLA binding pocket. Through examples, including the binding of abacavir and flucloxacillin to HLA-B*57:01, carbamazepine to HLA-B*15:02, and allopurinol to HLA-B*58:01, we demonstrate how docking analyses can provide insights into the drug and HLA allele-specificity of adverse events. Furthermore, the use of molecular docking to screen drugs with unknown IDR liability is examined, targeting either multiple HLA variants or a single specific variant. Despite multiple challenges, molecular docking presents a promising toolkit for investigating drug-HLA interactions and understanding IDR mechanisms, with significant implications for preemptive HLA typing and safer drug development.

Epitope mapping via in vitro deep mutational scanning methods and its applications

Epitope mapping is a technique employed to define the region of an antigen that elicits an immune response, providing crucial insight into the structural architecture of the antigen as well as epitope-paratope interactions. With this breadth of knowledge, immunotherapies, diagnostics, and vaccines are being developed with a rational and data-supported design. Traditional epitope mapping methods are laborious, time-intensive, and often lack the ability to screen proteins in a high-throughput manner or provide high resolution. Deep mutational scanning (DMS), however, is revolutionizing the field as it can screen all possible single amino acid mutations and provide an efficient and high-throughput way to infer the structures of both linear and three-dimensional epitopes with high resolution. Currently, more than 50 publications take this approach to efficiently identify enhancing or escaping mutations, with many then employing this information to rapidly develop broadly neutralizing antibodies, T-cell immunotherapies, vaccine platforms, or diagnostics. We provide a comprehensive review of the approaches to accomplish epitope mapping while also providing a summation of the development of DMS technology and its impactful applications.

Human microbiome-derived peptide affects the development of experimental autoimmune encephalomyelitis via molecular mimicry

BACKGROUND

Gut commensal microbiota has been identified as a potential environmental risk factor for multiple sclerosis (MS), and numerous studies have linked the commensal microorganism with the onset of MS. However, little is known about the mechanisms underlying the gut microbiome and host-immune system interaction.

METHODS

We employed bioinformatics methodologies to identify human microbial-derived peptides by analyzing their similarity to the MHC II-TCR binding patterns of self-antigens. Subsequently, we conducted a range of in vitro and in vivo assays to assess the encephalitogenic potential of these microbial-derived peptides.

FINDINGS

We analyzed 304,246 human microbiome genomes and 103 metagenomes collected from the MS cohort and identified 731 nonredundant analogs of myelin oligodendrocyte glycoprotein peptide 35-55 (MOG35-55). Of note, half of these analogs could bind to MHC II and interact with TCR through structural modeling of the interaction using fine-tuned AlphaFold. Among the 8 selected peptides, the peptide (P3) shows the ability to activate MOG35-55-specific CD4+ T cells in vitro. Furthermore, P3 shows encephalitogenic capacity and has the potential to induce EAE in some animals. Notably, mice immunized with a combination of P3 and MOG35-55 develop severe EAE. Additionally, dendritic cells could process and present P3 to MOG35-55-specific CD4+ T cells and activate these cells.

INTERPRETATION

Our data suggests the potential involvement of a MOG35-55-mimic peptide derived from the gut microbiota as a molecular trigger of EAE pathogenesis. Our findings offer direct evidence of how microbes can initiate the development of EAE, suggesting a potential explanation for the correlation between certain gut microorganisms and MS prevalence.

FUNDING

National Natural Science Foundation of China (82371350 to GY).

RankMHC: Learning to Rank Class-I Peptide-MHC Structural Models

The binding of peptides to class-I Major Histocompability Complex (MHC) receptors and their subsequent recognition downstream by T-cell receptors are crucial processes for most multicellular organisms to be able to fight various diseases. Thus, the identification of peptide antigens that can elicit an immune response is of immense importance for developing successful therapies for bacterial and viral infections, even cancer. Recently, studies have demonstrated the importance of peptide-MHC (pMHC) structural analysis, with pMHC structural modeling methods gradually becoming more popular in peptide antigen identification workflows. Most of the pMHC structural modeling tools provide an ensemble of candidate peptide poses in the MHC-I cleft, each associated with a score stemming from a scoring function, with the top scoring pose assumed to be the most representative of the ensemble. However, identifying the binding mode, that is, the peptide pose from the ensemble that is closer to an unavailable native structure, is not trivial. Oftentimes, the peptide poses characterized as best by a protein-ligand scoring function are not the ones that are the most representative of the actual structure. In this work, we frame the peptide binding pose identification problem as a Learning-to-Rank (LTR) problem. We present RankMHC, an LTR-based pMHC binding mode identification predictor, which is specifically trained to predict the most accurate ranking of an ensemble of pMHC conformations. RankMHC outperforms classical peptide-ligand scoring functions, as well as previous Machine Learning (ML)-based binding pose predictors. We further demonstrate that RankMHC can be used with many pMHC structural modeling tools that use different structural modeling protocols.

MHC-Fine: Fine-tuned AlphaFold for precise MHC-peptide complex prediction

The precise prediction of major histocompatibility complex (MHC)-peptide complex structures is pivotal for understanding cellular immune responses and advancing vaccine design. In this study, we enhanced AlphaFold's capabilities by fine-tuning it with a specialized dataset consisting of exclusively high-resolution class I MHC-peptide crystal structures. This tailored approach aimed to address the generalist nature of AlphaFold's original training, which, while broad-ranging, lacked the granularity necessary for the high-precision demands of class I MHC-peptide interaction prediction. A comparative analysis was conducted against the homology-modeling-based method Pandora as well as the AlphaFold multimer model. Our results demonstrate that our fine-tuned model outperforms others in terms of root-mean-square deviation (median value for Cα atoms for peptides is 0.66 Å) and also provides enhanced predicted local distance difference test scores, offering a more reliable assessment of the predicted structures. These advances have substantial implications for computational immunology, potentially accelerating the development of novel therapeutics and vaccines by providing a more precise computational lens through which to view MHC-peptide interactions.

The power and pitfalls of AlphaFold2 for structure prediction beyond rigid globular proteins

Artificial intelligence-driven advances in protein structure prediction in recent years have raised the question: has the protein structure-prediction problem been solved? Here, with a focus on nonglobular proteins, we highlight the many strengths and potential weaknesses of DeepMind's AlphaFold2 in the context of its biological and therapeutic applications. We summarize the subtleties associated with evaluation of AlphaFold2 model quality and reliability using the predicted local distance difference test (pLDDT) and predicted aligned error (PAE) values. We highlight various classes of proteins that AlphaFold2 can be applied to and the caveats involved. Concrete examples of how AlphaFold2 models can be integrated with experimental data in the form of small-angle X-ray scattering (SAXS), solution NMR, cryo-electron microscopy (cryo-EM) and X-ray diffraction are discussed. Finally, we highlight the need to move beyond structure prediction of rigid, static structural snapshots toward conformational ensembles and alternate biologically relevant states. The overarching theme is that careful consideration is due when using AlphaFold2-generated models to generate testable hypotheses and structural models, rather than treating predicted models as de facto ground truth structures.

A Computational Pipeline for Accurate Prioritization of Protein-Protein Binding Candidates in High-Throughput Protein Libraries

Identifying the interactome for a protein of interest is challenging due to the large number of possible binders. High-throughput experimental approaches narrow down possible binding partners but often include false positives. Furthermore, they provide no information about what the binding region is (e.g., the binding epitope). We introduce a novel computational pipeline based on an AlphaFold2 (AF) Competitive Binding Assay (AF-CBA) to identify proteins that bind a target of interest from a pull-down experiment and the binding epitope. Our focus is on proteins that bind the Extraterminal (ET) domain of Bromo and Extraterminal domain (BET) proteins, but we also introduce nine additional systems to show transferability to other peptide-protein systems. We describe a series of limitations to the methodology based on intrinsic deficiencies of AF and AF-CBA to help users identify scenarios where the approach will be most useful. Given the method's speed and accuracy, we anticipate its broad applicability to identify binding epitope regions among potential partners, setting the stage for experimental verification.

A structure-guided approach to predict MHC-I restriction of T cell receptors for public antigens

Peptides presented by class I major histocompatibility complex (MHC-I) proteins provide biomarkers for therapeutic targeting using T cell receptors (TCRs), TCR-mimicking antibodies (TMAs), or other engineered protein binders. Despite the extreme sequence diversity of the Human Leucocyte Antigen (HLA, the human MHC), a given TCR or TMA is restricted to recognize epitopic peptides in the context of a limited set of different HLA allotypes. Here, guided by our analysis of 96 TCR:pHLA complex structures in the Protein Data Bank (PDB), we identify TCR contact residues and classify 148 common HLA allotypes into T-cell cross-reactivity groups (T-CREGs) on the basis of their interaction surface features. Insights from our work have actionable value for resolving MHC-I restriction of TCRs, guiding therapeutic expansion of existing therapies, and informing the selection of peptide targets for forthcoming immunotherapy modalities.

A journey to your self: The vague definition of immune self and its practical implications

The identification of immunogenic peptides has become essential in an increasing number of fields in immunology, ranging from tumor immunotherapy to vaccine development. The nature of the adaptive immune response is shaped by the similarity between foreign and self-protein sequences, a concept extensively applied in numerous studies. Can we precisely define the degree of similarity to self? Furthermore, do we accurately define immune self? In the current work, we aim to unravel the conceptual and mechanistic vagueness hindering the assessment of self-similarity. Accordingly, we demonstrate the remarkably low consistency among commonly employed measures and highlight potential avenues for future research.

Can AlphaFold's breakthrough in protein structure help decode the fundamental principles of adaptive cellular immunity?

T cells are essential immune cells responsible for identifying and eliminating pathogens. Through interactions between their T-cell antigen receptors (TCRs) and antigens presented by major histocompatibility complex molecules (MHCs) or MHC-like molecules, T cells discriminate foreign and self peptides. Determining the fundamental principles that govern these interactions has important implications in numerous medical contexts. However, reconstructing a map between T cells and their antagonist antigens remains an open challenge for the field of immunology, and success of in silico reconstructions of this relationship has remained incremental. In this Perspective, we discuss the role that new state-of-the-art deep-learning models for predicting protein structure may play in resolving some of the unanswered questions the field faces linking TCR and peptide-MHC properties to T-cell specificity. We provide a comprehensive overview of structural databases and the evolution of predictive models, and highlight the breakthrough AlphaFold provided the field.

The impact of SARS-CoV-2 spike mutation on peptide presentation is HLA allomorph-specific

CD8+ T cells are crucial for viral elimination and recovery from viral infection. Nonetheless, the current understanding of the T cell response to SARS-CoV-2 at the antigen level remains limited. The Spike protein is an external structural protein that is prone to mutations, threatening the efficacy of current vaccines. Therefore, we have characterised the immune response towards the immunogenic Spike-derived peptide (S976-984, VLNDILSRL), restricted to the HLA-A*02:01 molecule, which is mutated in both Alpha (S982A) and Omicron BA.1 (L981F) variants of concern. We determined that the mutation in the Alpha variant (S982A) impacted both the stability and conformation of the peptide, bound to HLA-A*02:01, in comparison to the original S976-984. We identified a longer and overlapping immunogenic peptide (S975-984, SVLNDILSRL) that could be presented by HLA-A*02:01, HLA-A*11:01 and HLA-B*13:01 allomorphs. We showed that S975-specific CD8+ T cells were weakly cross-reactive to the mutant peptides despite their similar conformations when presented by HLA-A*11:01. Altogether, our results show that the impact of SARS-CoV-2 mutations on peptide presentation is HLA allomorph-specific, and that post vaccination there are T cells able to react and cross-react towards the variant of concern peptides.