T Cell Epitope Predictions.

Throughout the body, T cells monitor MHC-bound ligands expressed on the surface of essentially all cell types. MHC ligands that trigger a T cell immune response are referred to as T cell epitopes. Identifying such epitopes enables tracking, phenotyping, and stimulating T cells involved in immune responses in infectious disease, allergy, autoimmunity, transplantation, and cancer. The specific T cell epitopes recognized in an individual are determined by genetic factors such as the MHC molecules the individual expresses, in parallel to the individual's environmental exposure history. The complexity and importance of T cell epitope mapping have motivated the development of computational approaches that predict what T cell epitopes are likely to be recognized in a given individual or in a broader population. Such predictions guide experimental epitope mapping studies and enable computational analysis of the immunogenic potential of a given protein sequence region.

The Art and Science of Molecular Docking.

Molecular docking has become an essential part of a structural biologist's and medicinal chemist's toolkits. Given a chemical compound and the three-dimensional structure of a molecular target-for example, a protein-docking methods fit the compound into the target, predicting the compound's bound structure and binding energy. Docking can be used to discover novel ligands for a target by screening large virtual compound libraries. Docking can also provide a useful starting point for structure-based ligand optimization or for investigating a ligand's mechanism of action. Advances in computational methods, including both physics-based and machine learning approaches, as well as in complementary experimental techniques, are making docking an even more powerful tool. We review how docking works and how it can drive drug discovery and biological research. We also describe its current limitations and ongoing efforts to overcome them.

Toward a foundation model of causal cell and tissue biology with a Perturbation Cell and Tissue Atlas.

Comprehensively charting the biologically causal circuits that govern the phenotypic space of human cells has often been viewed as an insurmountable challenge. However, in the last decade, a suite of interleaved experimental and computational technologies has arisen that is making this fundamental goal increasingly tractable. Pooled CRISPR-based perturbation screens with high-content molecular and/or image-based readouts are now enabling researchers to probe, map, and decipher genetically causal circuits at increasing scale. This scale is now eminently suitable for the deployment of artificial intelligence and machine learning (AI/ML) to both direct further experiments and to predict or generate information that was not-and sometimes cannot-be gathered experimentally. By combining and iterating those through experiments that are designed for inference, we now envision a Perturbation Cell Atlas as a generative causal foundation model to unify human cell biology.

Machine learning interpretable models of cell mechanics from protein images.

Cellular form and function emerge from complex mechanochemical systems within the cytoplasm. Currently, no systematic strategy exists to infer large-scale physical properties of a cell from its molecular components. This is an obstacle to understanding processes such as cell adhesion and migration. Here, we develop a data-driven modeling pipeline to learn the mechanical behavior of adherent cells. We first train neural networks to predict cellular forces from images of cytoskeletal proteins. Strikingly, experimental images of a single focal adhesion (FA) protein, such as zyxin, are sufficient to predict forces and can generalize to unseen biological regimes. Using this observation, we develop two approaches-one constrained by physics and the other agnostic-to construct data-driven continuum models of cellular forces. Both reveal how cellular forces are encoded by two distinct length scales. Beyond adherent cell mechanics, our work serves as a case study for integrating neural networks into predictive models for cell biology.

Discovery of antimicrobial peptides in the global microbiome with machine learning.

Novel antibiotics are urgently needed to combat the antibiotic-resistance crisis. We present a machine-learning-based approach to predict antimicrobial peptides (AMPs) within the global microbiome and leverage a vast dataset of 63,410 metagenomes and 87,920 prokaryotic genomes from environmental and host-associated habitats to create the AMPSphere, a comprehensive catalog comprising 863,498 non-redundant peptides, few of which match existing databases. AMPSphere provides insights into the evolutionary origins of peptides, including by duplication or gene truncation of longer sequences, and we observed that AMP production varies by habitat. To validate our predictions, we synthesized and tested 100 AMPs against clinically relevant drug-resistant pathogens and human gut commensals both in vitro and in vivo. A total of 79 peptides were active, with 63 targeting pathogens. These active AMPs exhibited antibacterial activity by disrupting bacterial membranes. In conclusion, our approach identified nearly one million prokaryotic AMP sequences, an open-access resource for antibiotic discovery.

From patterns to patients: Advances in clinical machine learning for cancer diagnosis, prognosis, and treatment.

Machine learning (ML) is increasingly used in clinical oncology to diagnose cancers, predict patient outcomes, and inform treatment planning. Here, we review recent applications of ML across the clinical oncology workflow. We review how these techniques are applied to medical imaging and to molecular data obtained from liquid and solid tumor biopsies for cancer diagnosis, prognosis, and treatment design. We discuss key considerations in developing ML for the distinct challenges posed by imaging and molecular data. Finally, we examine ML models approved for cancer-related patient usage by regulatory agencies and discuss approaches to improve the clinical usefulness of ML.

Designing antibodies as therapeutics.

Antibody therapeutics are a large and rapidly expanding drug class providing major health benefits. We provide a snapshot of current antibody therapeutics including their formats, common targets, therapeutic areas, and routes of administration. Our focus is on selected emerging directions in antibody design where progress may provide a broad benefit. These topics include enhancing antibodies for cancer, antibody delivery to organs such as the brain, gastrointestinal tract, and lungs, plus antibody developability challenges including immunogenicity risk assessment and mitigation and subcutaneous delivery. Machine learning has the potential, albeit as yet largely unrealized, for a transformative future impact on antibody discovery and engineering.

A blood atlas of COVID-19 defines hallmarks of disease severity and specificity.

Treatment of severe COVID-19 is currently limited by clinical heterogeneity and incomplete description of specific immune biomarkers. We present here a comprehensive multi-omic blood atlas for patients with varying COVID-19 severity in an integrated comparison with influenza and sepsis patients versus healthy volunteers. We identify immune signatures and correlates of host response. Hallmarks of disease severity involved cells, their inflammatory mediators and networks, including progenitor cells and specific myeloid and lymphocyte subsets, features of the immune repertoire, acute phase response, metabolism, and coagulation. Persisting immune activation involving AP-1/p38MAPK was a specific feature of COVID-19. The plasma proteome enabled sub-phenotyping into patient clusters, predictive of severity and outcome. Systems-based integrative analyses including tensor and matrix decomposition of all modalities revealed feature groupings linked with severity and specificity compared to influenza and sepsis. Our approach and blood atlas will support future drug development, clinical trial design, and personalized medicine approaches for COVID-19.

How Machine Learning Will Transform Biomedicine.

This Perspective explores the application of machine learning toward improved diagnosis and treatment. We outline a vision for how machine learning can transform three broad areas of biomedicine: clinical diagnostics, precision treatments, and health monitoring, where the goal is to maintain health through a range of diseases and the normal aging process. For each area, early instances of successful machine learning applications are discussed, as well as opportunities and challenges for machine learning. When these challenges are met, machine learning promises a future of rigorous, outcomes-based medicine with detection, diagnosis, and treatment strategies that are continuously adapted to individual and environmental differences.

Determinants of Base Editing Outcomes from Target Library Analysis and Machine Learning.

Although base editors are widely used to install targeted point mutations, the factors that determine base editing outcomes are not well understood. We characterized sequence-activity relationships of 11 cytosine and adenine base editors (CBEs and ABEs) on 38,538 genomically integrated targets in mammalian cells and used the resulting outcomes to train BE-Hive, a machine learning model that accurately predicts base editing genotypic outcomes (R ≈ 0.9) and efficiency (R ≈ 0.7). We corrected 3,388 disease-associated SNVs with ≥90% precision, including 675 alleles with bystander nucleotides that BE-Hive correctly predicted would not be edited. We discovered determinants of previously unpredictable C-to-G, or C-to-A editing and used these discoveries to correct coding sequences of 174 pathogenic transversion SNVs with ≥90% precision. Finally, we used insights from BE-Hive to engineer novel CBE variants that modulate editing outcomes. These discoveries illuminate base editing, enable editing at previously intractable targets, and provide new base editors with improved editing capabilities.

An Amygdala Circuit Mediates Experience-Dependent Momentary Arrests during Exploration.

Exploration of novel environments ensures survival and evolutionary fitness. It is expressed through exploratory bouts and arrests that change dynamically based on experience. Neural circuits mediating exploratory behavior should therefore integrate experience and use it to select the proper behavioral output. Using a spatial exploration assay, we uncovered an experience-dependent increase in momentary arrests in locations where animals arrested previously. Calcium imaging in freely exploring mice revealed a genetically and projection-defined neuronal ensemble in the basolateral amygdala that is active during self-paced behavioral arrests. This ensemble was recruited in an experience-dependent manner, and closed-loop optogenetic manipulation of these neurons revealed that they are sufficient and necessary to drive experience-dependent arrests during exploration. Projection-specific imaging and optogenetic experiments revealed that these arrests are effected by basolateral amygdala neurons projecting to the central amygdala, uncovering an amygdala circuit that mediates momentary arrests in familiar places but not avoidance or anxiety/fear-like behaviors.

Proteomic and Metabolomic Characterization of COVID-19 Patient Sera.

Early detection and effective treatment of severe COVID-19 patients remain major challenges. Here, we performed proteomic and metabolomic profiling of sera from 46 COVID-19 and 53 control individuals. We then trained a machine learning model using proteomic and metabolomic measurements from a training cohort of 18 non-severe and 13 severe patients. The model was validated using 10 independent patients, 7 of which were correctly classified. Targeted proteomics and metabolomics assays were employed to further validate this molecular classifier in a second test cohort of 19 COVID-19 patients, leading to 16 correct assignments. We identified molecular changes in the sera of COVID-19 patients compared to other groups implicating dysregulation of macrophage, platelet degranulation, complement system pathways, and massive metabolic suppression. This study revealed characteristic protein and metabolite changes in the sera of severe COVID-19 patients, which might be used in selection of potential blood biomarkers for severity evaluation.

Directed Evolution of a Selective and Sensitive Serotonin Sensor via Machine Learning.

Serotonin plays a central role in cognition and is the target of most pharmaceuticals for psychiatric disorders. Existing drugs have limited efficacy; creation of improved versions will require better understanding of serotonergic circuitry, which has been hampered by our inability to monitor serotonin release and transport with high spatial and temporal resolution. We developed and applied a binding-pocket redesign strategy, guided by machine learning, to create a high-performance, soluble, fluorescent serotonin sensor (iSeroSnFR), enabling optical detection of millisecond-scale serotonin transients. We demonstrate that iSeroSnFR can be used to detect serotonin release in freely behaving mice during fear conditioning, social interaction, and sleep/wake transitions. We also developed a robust assay of serotonin transporter function and modulation by drugs. We expect that both machine-learning-guided binding-pocket redesign and iSeroSnFR will have broad utility for the development of other sensors and in vitro and in vivo serotonin detection, respectively.

Extracellular Vesicle and Particle Biomarkers Define Multiple Human Cancers.

There is an unmet clinical need for improved tissue and liquid biopsy tools for cancer detection. We investigated the proteomic profile of extracellular vesicles and particles (EVPs) in 426 human samples from tissue explants (TEs), plasma, and other bodily fluids. Among traditional exosome markers, CD9, HSPA8, ALIX, and HSP90AB1 represent pan-EVP markers, while ACTB, MSN, and RAP1B are novel pan-EVP markers. To confirm that EVPs are ideal diagnostic tools, we analyzed proteomes of TE- (n = 151) and plasma-derived (n = 120) EVPs. Comparison of TE EVPs identified proteins (e.g., VCAN, TNC, and THBS2) that distinguish tumors from normal tissues with 90% sensitivity/94% specificity. Machine-learning classification of plasma-derived EVP cargo, including immunoglobulins, revealed 95% sensitivity/90% specificity in detecting cancer. Finally, we defined a panel of tumor-type-specific EVP proteins in TEs and plasma, which can classify tumors of unknown primary origin. Thus, EVP proteins can serve as reliable biomarkers for cancer detection and determining cancer type.

A Deep Learning Approach to Antibiotic Discovery.

Due to the rapid emergence of antibiotic-resistant bacteria, there is a growing need to discover new antibiotics. To address this challenge, we trained a deep neural network capable of predicting molecules with antibacterial activity. We performed predictions on multiple chemical libraries and discovered a molecule from the Drug Repurposing Hub-halicin-that is structurally divergent from conventional antibiotics and displays bactericidal activity against a wide phylogenetic spectrum of pathogens including Mycobacterium tuberculosis and carbapenem-resistant Enterobacteriaceae. Halicin also effectively treated Clostridioides difficile and pan-resistant Acinetobacter baumannii infections in murine models. Additionally, from a discrete set of 23 empirically tested predictions from >107 million molecules curated from the ZINC15 database, our model identified eight antibacterial compounds that are structurally distant from known antibiotics. This work highlights the utility of deep learning approaches to expand our antibiotic arsenal through the discovery of structurally distinct antibacterial molecules.

A guide to systems-level immunomics.

The immune system is highly complex and distributed throughout an organism, with hundreds to thousands of cell states existing in parallel with diverse molecular pathways interacting in a highly dynamic and coordinated fashion. Although the characterization of individual genes and molecules is of the utmost importance for understanding immune-system function, high-throughput, high-resolution omics technologies combined with sophisticated computational modeling and machine-learning approaches are creating opportunities to complement standard immunological methods with new insights into immune-system dynamics. Like systems immunology itself, immunology researchers must take advantage of these technologies and form their own diverse networks, connecting with researchers from other disciplines. This Review is an introduction and 'how-to guide' for immunologists with no particular experience in the field of omics but with the intention to learn about and apply these systems-level approaches, and for immunologists who want to make the most of interdisciplinary networks.

A guide to machine learning for biologists.

The expanding scale and inherent complexity of biological data have encouraged a growing use of machine learning in biology to build informative and predictive models of the underlying biological processes. All machine learning techniques fit models to data; however, the specific methods are quite varied and can at first glance seem bewildering. In this Review, we aim to provide readers with a gentle introduction to a few key machine learning techniques, including the most recently developed and widely used techniques involving deep neural networks. We describe how different techniques may be suited to specific types of biological data, and also discuss some best practices and points to consider when one is embarking on experiments involving machine learning. Some emerging directions in machine learning methodology are also discussed.

Antibiotic Lethality Is Impacted by Nutrient Availabilities: New Insights from Machine Learning.

In this issue of Cell, Yang, Wright et al. describe a machine learning approach that that can provide mechanistic insight from chemical screens. They use this approach to uncover how the nutritional availability for Escherichia coli impacts lethality toward three widely used antibiotics.

Inherited and De Novo Genetic Risk for Autism Impacts Shared Networks.

We performed a comprehensive assessment of rare inherited variation in autism spectrum disorder (ASD) by analyzing whole-genome sequences of 2,308 individuals from families with multiple affected children. We implicate 69 genes in ASD risk, including 24 passing genome-wide Bonferroni correction and 16 new ASD risk genes, most supported by rare inherited variants, a substantial extension of previous findings. Biological pathways enriched for genes harboring inherited variants represent cytoskeletal organization and ion transport, which are distinct from pathways implicated in previous studies. Nevertheless, the de novo and inherited genes contribute to a common protein-protein interaction network. We also identified structural variants (SVs) affecting non-coding regions, implicating recurrent deletions in the promoters of DLG2 and NR3C2. Loss of nr3c2 function in zebrafish disrupts sleep and social function, overlapping with human ASD-related phenotypes. These data support the utility of studying multiplex families in ASD and are available through the Hartwell Autism Research and Technology portal.

A White-Box Machine Learning Approach for Revealing Antibiotic Mechanisms of Action.

Current machine learning techniques enable robust association of biological signals with measured phenotypes, but these approaches are incapable of identifying causal relationships. Here, we develop an integrated "white-box" biochemical screening, network modeling, and machine learning approach for revealing causal mechanisms and apply this approach to understanding antibiotic efficacy. We counter-screen diverse metabolites against bactericidal antibiotics in Escherichia coli and simulate their corresponding metabolic states using a genome-scale metabolic network model. Regression of the measured screening data on model simulations reveals that purine biosynthesis participates in antibiotic lethality, which we validate experimentally. We show that antibiotic-induced adenine limitation increases ATP demand, which elevates central carbon metabolism activity and oxygen consumption, enhancing the killing effects of antibiotics. This work demonstrates how prospective network modeling can couple with machine learning to identify complex causal mechanisms underlying drug efficacy.