Biology subjects

Humphreys, I. R.

Publications and source records attributed to Humphreys, I. R..

13 recordsLinked to original sources

bioRxiv2026

In vivo screening reveals new regulators of Natural Killer cell development and functional response to acute cytomegalovirus infection

Natural killer (NK) cells are critical innate immune effectors in antiviral and antitumour responses. However, the factors governing NK cell development and function remain incompletely understood. To address this, we adopted an in vivo screening approach, generating a panel of mice deficient in genes differentially expressed across NK cell maturation stages or following viral infection. We performed a combinational screen comprising baseline immunophenotyping and mouse cytomegalovirus (MCMV) challenge. We identified novel regulators of NK cell development, including transcriptional regulators and proteins with putative trafficking functions. MCMV challenge studies additionally identified proteins that impacted antiviral immunity independently of NK cell phenotypes, including a novel regulator of NK cell degranulation, Synaptotagmin-like protein 3. Identified genes exhibited reduced tolerance to loss-of-function in large scale human sequencing studies and evidence for reduced NK cell numbers in a cohort of immunodeficiency patients. Together, these findings provide a resource of NK-expressed genes important for NK cell maturation and function, with relevance to human health. SummaryPring et al. conduct an in vivo screen in mice identifying novel regulators of natural killer (NK) cell development and/or antiviral functionality. Large-scale human sequencing and human immunodeficiency patients revealed evidence for the relevance of identified genes in human health.

immunology↗

bioRxiv2025

Accelerating Biomolecular Modeling with AtomWorks andRF3

Deep learning methods trained on protein structure databases have revolutionized biomolecular structure prediction, but developing and training new models remains a considerable challenge. To facilitate the development of new models, we present AtomWorks: a broadly applicable data framework for developing state-of-the-art biomolecular foundation models spanning diverse tasks, including structure prediction, generative protein design, and fixed backbone sequence design. We use AtomWorks to train RosettaFold-3 (RF3), a structure prediction network capable of predicting arbitrary biomolecular complexes with an improved treatment of chirality that narrows the performance gap between closed-source AlphaFold3 (AF3) and existing open-source implementations. We expect that AtomWorks will accelerate the next generation of open-source biomolecular machine learning models and that RF3 will be broadly useful as a structure prediction tool. To this end, we release the AtomWorks framework (https://github.com/RosettaCommons/atomworks), together with curated training data, code and model weights for RF3 (https://github.com/RosettaCommons/modelforge) under a permissive BSD license.

biochemistry↗

bioRxiv2025

Cytomegalovirus-induced T cell responses accelerate Alzheimer's disease progression in mice

Infections have long been implicated as causative factors in Alzheimers disease (AD). Multiple studies have further suggested a key role for herpesviruses, such as cytomegalovirus (CMV). Using transgenic 3xTg-AD mice, we demonstrate that systemic infection with the {beta}-herpesvirus murine cytomegalovirus (MCMV) accelerates the development of cognitive decline, tauopathy, and synaptic loss in the hippocampus, all of which are key features of AD. Accelerated disease progression after infection was associated with substantial lymphocyte infiltration into the brain dominated by CD8+ T cells specific for MCMV. Moreover, T cell receptor analyses revealed that these responses were clonally diverse, suggesting that multiple viral antigens were targeted in the brain during chronic infection with MCMV. T cell depletion during virus chronicity rescued infection-induced cognitive decline. In addition, antiviral drug treatment reduced lymphocytic infiltrates in the brain and reversed cognitive decline, suggesting potential clinical utility. These data provide a mechanistic link between chronic viral infections and the development of AD.

immunology↗

bioRxiv2024

Computing the Human Interactome

Protein-protein interactions (PPI) are essential for biological function. Recent advances in coevolutionary analysis and Deep Learning (DL) based protein structure prediction have enabled comprehensive PPI identification in bacterial and yeast proteomes, but these approaches have limited success to date for the more complex human proteome. Here, we overcome this challenge by 1) enhancing the coevolutionary signals with 7-fold deeper multiple sequence alignments harvested from 30 petabytes of unassembled genomic data, and 2) developing a new DL network trained on augmented datasets of domain-domain interactions from 200 million predicted protein structures. These advancements allow us to systematically screen through 200 million human protein pairs and predict 18,316 PPIs with an expected precision of 90%, among which 5,578 are novel predictions. 3D models of these predicted PPIs nearly triple the number of human PPIs with accurate structural information, providing numerous insights into protein function and mechanisms of human diseases.

biochemistry↗

bioRxiv2024

Computational design of serine hydrolases

Enzymes that proceed through multistep reaction mechanisms often utilize complex, polar active sites positioned with sub-angstrom precision to mediate distinct chemical steps, which makes their de novo construction extremely challenging. We sought to overcome this challenge using the classic catalytic triad and oxyanion hole of serine hydrolases as a model system. We used RFdiffusion1 to generate proteins housing catalytic sites of increasing complexity and varying geometry, and a newly developed ensemble generation method called ChemNet to assess active site geometry and preorganization at each step of the reaction. Experimental characterization revealed novel serine hydrolases that catalyze ester hydrolysis with catalytic efficiencies (kcat/Km) up to 3.8 x 103 M-1 s-1, closely match the design models (C RMSDs < 1 [A]), and have folds distinct from natural serine hydrolases. In silico selection of designs based on active site preorganization across the reaction coordinate considerably increased success rates, enabling identification of new catalysts in screens of as few as 20 designs. Our de novo buildup approach provides insight into the geometric determinants of catalysis that complements what can be obtained from structural and mutational studies of native enzymes (in which catalytic group geometry and active site makeup cannot be so systematically varied), and provides a roadmap for the design of industrially relevant serine hydrolases and, more generally, for designing complex enzymes that catalyze multi-step transformations.

biochemistry↗

bioRxiv2024

Essential and virulence-related protein interactions of pathogens revealed through deep learning

Identification of bacterial protein-protein interactions and predicting the structures of the complexes could aid in the understanding of pathogenicity mechanisms and developing treatments for infectious diseases. Here, we developed a deep learning-based pipeline that leverages residue-residue coevolution and protein structure prediction to systematically identify and structurally characterize protein-protein interactions at the proteome-wide scale. Using this pipeline, we searched through 78 million pairs of proteins across 19 human bacterial pathogens and identified 1923 confidently predicted complexes involving essential genes and 256 involving virulence factors. Many of these complexes were not previously known; we experimentally tested 12 such predictions, and half of them were validated. The predicted interactions span core metabolic and virulence pathways ranging from post-transcriptional modification to acid neutralization to outer membrane machinery and should contribute to our understanding of the biology of these important pathogens and the design of drugs to combat them.

biochemistry↗

bioRxiv2023

Generalized Biomolecular Modeling and Design with RoseTTAFold All-Atom

Although AlphaFold2 (AF2) and RoseTTAFold (RF) have transformed structural biology by enabling high-accuracy protein structure modeling, they are unable to model covalent modifications or interactions with small molecules and other non-protein molecules that can play key roles in biological function. Here, we describe RoseTTAFold All-Atom (RFAA), a deep network capable of modeling full biological assemblies containing proteins, nucleic acids, small molecules, metals, and covalent modifications given the sequences of the polymers and the atomic bonded geometry of the small molecules and covalent modifications. Following training on structures of full biological assemblies in the Protein Data Bank (PDB), RFAA has comparable protein structure prediction accuracy to AF2, excellent performance in CAMEO for flexible backbone small molecule docking, and reasonable prediction accuracy for protein covalent modifications and assemblies of proteins with multiple nucleic acid chains and small molecules which, to our knowledge, no existing method can model simultaneously. By fine-tuning on diffusive denoising tasks, we develop RFdiffusion All-Atom (RFdiffusionAA), which generates binding pockets by directly building protein structures around small molecules and other non-protein molecules. Starting from random distributions of amino acid residues surrounding target small molecules, we design and experimentally validate proteins that bind the cardiac disease therapeutic digoxigenin, the enzymatic cofactor heme, and optically active bilin molecules with potential for expanding the range of wavelengths captured by photosynthesis. We anticipate that RFAA and RFdiffusionAA will be widely useful for modeling and designing complex biomolecular systems.

biochemistry↗

bioRxiv2023

Genetic manipulation of candidate phyla radiation bacteria provides functional insights into microbial dark matter

The study of bacteria has yielded fundamental insights into cellular biology and physiology, biotechnological advances and many therapeutics. Yet due to a lack of suitable tools, the significant portion of bacterial diversity held within the candidate phyla radiation (CPR) remains inaccessible to such pursuits. Here we show that CPR bacteria belonging to the phylum Saccharibacteria exhibit natural competence. We exploit this property to develop methods for their genetic manipulation, including the insertion of heterologous sequences and the construction of targeted gene deletions. Imaging of fluorescent protein-labeled Saccharibacteria provides high spatiotemporal resolution of phenomena accompanying epibiotic growth and a transposon insertion sequencing genome-wide screen reveals the contribution of enigmatic Saccharibacterial genes to growth on their Actinobacteria hosts. Finally, we leverage metagenomic data to provide cutting-edge protein structure-based bioinformatic resources that support the strain Southlakia epibionticum and its corresponding host, Actinomyces israelii, as a model system for unlocking the molecular underpinnings of the epibiotic lifestyle.

microbiology↗

bioRxiv2021

Structures of core eukaryotic protein complexes

Protein-protein interactions play critical roles in biology, but despite decades of effort, the structures of many eukaryotic protein complexes are unknown, and there are likely many interactions that have not yet been identified. Here, we take advantage of recent advances in proteome-wide amino acid coevolution analysis and deep-learning-based structure modeling to systematically identify and build accurate models of core eukaryotic protein complexes, as represented within the Saccharomyces cerevisiae proteome. We use a combination of RoseTTAFold and AlphaFold to screen through paired multiple sequence alignments for 8.3 million pairs of S. cerevisiae proteins and build models for strongly predicted protein assemblies with two to five components. Comparison to existing interaction and structural data suggests that these predictions are likely to be quite accurate. We provide structure models spanning almost all key processes in Eukaryotic cells for 104 protein assemblies which have not been previously identified, and 608 which have not been structurally characterized. One-sentence summaryWe take advantage of recent advances in proteome-wide amino acid coevolution analysis and deep-learning-based structure modeling to systematically identify and build accurate models of core eukaryotic protein complexes.

biochemistry↗

bioRxiv2021

ADNKA overcomes SARS-CoV2-mediated NK cell inhibition through non-spike antibodies

The outcome of infection is dependent on the ability of viruses to manipulate the infected cell to evade immunity, and the ability of the immune response to overcome this evasion. Understanding this process is key to understanding pathogenesis, genetic risk factors, and both natural and vaccine-induced immunity. SARS-CoV-2 antagonises the innate interferon response, but whether it manipulates innate cellular immunity is unclear. An unbiased proteomic analysis determined how cell surface protein expression is altered on SARS-CoV-2-infected lung epithelial cells, showing downregulation of activating NK ligands B7-H6, MICA, ULBP2, and Nectin1, with minimal effects on MHC-I. This correlated with a reduction in NK cell activation, identifying a novel mechanism by which SARS-CoV2 antagonises innate immunity. Later in the disease process, strong antibody-dependent NK cell activation (ADNKA) developed. These responses were sustained for at least 6 months in most patients, and led to high levels of pro-inflammatory cytokine production. Depletion of spike-specific antibodies confirmed their dominant role in neutralisation, but these antibodies played only a minor role in ADNKA compared to antibodies to other proteins, including ORF3a, Membrane, and Nucleocapsid. In contrast, ADNKA induced following vaccination was focussed solely on spike, was weaker than ADNKA following natural infection, and was not boosted by the second dose. These insights have important implications for understanding disease progression, vaccine efficacy, and vaccine design.

immunology↗

bioRxiv2020

Optimal CD8+ T-cell memory formation following subcutaneous cytomegalovirus infection requires virus replication but not early dendritic cell responses

Cytomegalovirus (CMV) induction of large frequencies of highly functional memory T-cells has attracted much interest in the utility of CMV-based vaccine vectors, with exciting preclinical data obtained in models of infectious diseases and cancer. However, pathogenesis of human CMV (HCMV) remains a concern. Attenuated CMV-based vectors, such as replication- or spread-deficient viruses potentially offer an alternative to fully replicating vectors. However, it is not well-understood how CMV attenuation impacts vector immunogenicity, in particularly when administered via relevant routes of immunization such as the skin. Herein we used the murine cytomegalovirus (MCMV) model to investigate the impact of vector attenuation on T-cell memory formation following subcutaneous administration. We found that the spread deficient virus ({Delta}gL-MCMV) was impaired in its ability to induce memory CD8+ T-cells reactive to some (M38, IE1) but not all (IE3) viral antigens. Impaired memory T-cell development was associated with a preferential and pronounced loss of polyfunctional (IFN-{gamma}+ TNF-+) T-cells, and was not rescued by increasing the dose of replication-defective MCMV. Finally, whilst vector attenuation reduced dendritic cell (DC) recruitment to skindraining lymph nodes, systematic depletion of multiple DC subsets during acute subcutaneous MCMV infection had a negligible impact on T-cell memory formation, implying that attenuated responses induced by replication-deficient vectors were likely not a consequence of impaired initial DC activation. Thus, overall, these data imply that the choice of antigen and/or cloning strategy of exogenous antigen in combination with the route of immunization may influence the ability of attenuated CMV vectors to induce robust functional T-cell memory.

immunology↗

bioRxiv2020

Interspecies comparative metagenomics reveals correlated gut microbiome functional capacities among vertebrates

While recent research reveals that the gut microbiome drives vertebrate health, little is known about whether the mechanisms these microbes employ to interact with physiology are consistent across host species. To help close this knowledge gap, we compared gut metagenomes across 10 vertebrate species, including biomedical animal models, to define the inter-species variation in the biochemical pathways encoded by gut microbiota. Doing so revealed gut-enriched pathways conserved across vertebrates, as well as pathways that vary concordantly with host evolutionary history. Overall, the functional capacity of the non-human gut microbiome generally reflects that of humans, though a subset of the pathways encoded by human gut microbiota are not well represented in non-human microbiomes. Collectively, these results support the use of animal models to study the mechanisms through which gut microbes impact human health, but suggest that researchers should cautiously consider which model will optimally represent a specific mechanism of interest. SignificanceEfforts to understand how the gut microbiome interacts with human physiology frequently relies on the use of animal models. However, it is generally not understood if the biochemical pathways encoded in gut microbiomes of these different animal models - which define the routes of interaction between gut microbes and their hosts - reflect those found in the human gut. To address this question, we compared gut metagenomes generated 10 different vertebrate lineages. In so doing, our study revealed that non-human gut metagenomes generally encode a set of pathways that are consistent with those found in the human gut. However, some human metagenome pathways are poorly represented in non-human guts, including pathways implicated in disease. Moreover, our analysis identified pathways that appear to be conserved across vertebrates, as well as pathways that are linked to the evolutionary history of their hosts, observations that hold potential to clarify the basis for phylosymbiosis.

microbiology↗

bioRxiv2020

An integrated gene catalog of the zebrafish gut microbiome reveals significant homology with mammalian microbiomes

Gut microbiome research increasingly utilizes zebrafish (Danio rerio) given their amenability to high-throughput experimental designs. However, the utility of zebrafish for discerning translationally relevant host-microbiome interactions is constrained by a paucity of knowledge about the biological functions that zebrafish gut microbiota can execute, how these functions associate with zebrafish physiology, and the degree of homology between the genes encoded by the zebrafish and human gut microbiomes. To address this knowledge gap, we generated a foundational catalog of zebrafish gut microbiome genomic diversity consisting of 1,569,102 non-redundant genes from twenty-nine individual fish. We identified hundreds of novel microbial genes as well as dozens of biosynthetic gene clusters of potential clinical interest. The genomic diversity of the zebrafish gut microbiome varied significantly across diets and this variance associated with altered expression of intestinal genes involved in inflammation and immune activation. Zebrafish, mouse, and human fecal microbiomes shared > 50% of their total genomic diversity and the vast majority of gene family abundance for each individual metagenome ([~]99%) was accounted for by genes that comprised this shared fraction. These results indicate that the zebrafish gut houses a functionally diverse microbial community that manifests extensive homology to that of humans and mice despite substantial disparities in taxonomic composition. We anticipate that the gene catalog developed here will enable future mechanistic study of host-microbiome interactions using the zebrafish model. ImportanceZebrafish have emerged as an important model system for defining host-microbiome interactions. However, the utility of this model is blunted by limited insight into the functions that are carried by zebrafish gut microbiota, their relationship with zebrafish physiology, and their consistency with the functions carried by human gut microbiota. To address these limitations, we constructed the first genomic database of zebrafish gut microbiome diversity. We use this novel resource to demonstrate that the genomic diversity of the zebrafish gut microbiome varies with diet and this variance links with altered intestinal gene expression. We also identify substantial homology between zebrafish, human, and mouse metagenomic diversity, indicating that these microbiomes may operate similarly.

microbiology↗