Biology subjects

Nayfach, S.

Publications and source records attributed to Nayfach, S..

18 recordsLinked to original sources

bioRxiv2025

A genomic atlas of the human gut virome elucidates genetic factors shaping host interactions

Viruses are key modulators of human gut microbiome composition and function. While metagenomic sequencing has enabled culture-independent discovery of gut bacteriophage diversity, existing genomic catalogues suffer from limited geographic representation, sparse taxonomic classification, and insufficient functional annotation, hindering detailed investigation into phage biology. Here, we present the Unified Human Gastrointestinal Virome (UHGV), a collection of 873,994 viral genomes from globally diverse populations that addresses these limitations. UHGV provides high-quality virome references with extensive host predictions, comprehensive functional annotations, protein structures, a classification framework for comparative analysis, and a web portal to facilitate data access. Using UHGV to profile worldwide metagenomes, we found that host range breadth is strongly associated with phage prevalence. Additionally, we identified diversity-generating retroelements and DNA methyltransferases as key factors enabling phage populations to access diverse hosts, revealing how specific genomic features contribute to global phage distribution patterns. UHGV is available at http://uhgv.jgi.doe.gov.

genomics↗

bioRxiv2025

Small uncultured viral-like entities redraw the origin of viruses and cellular compartments

The HK97-fold is an ancient, conserved protein structure that forms protein shells. It is the building block of capsids that protect the genome of viruses in the Duplodnaviria realm, infecting organisms across all domains of life. It is also the building block of encapsulins, compartments that confine biochemical reactions in prokaryotes. Recent genomic studies have hypothesized that encapsulins evolved from HK97-fold viruses. However, this evolutionary pathway is challenging to justify biophysically because HK97-fold viruses form larger and more complex protein shells than encapsulins. We addressed this paradox by searching for smaller and simpler HK97-fold viral capsids across ecosystems. The investigation yielded a well-defined group of viral entities, which encode HK97-fold proteins displaying molecular similarities with encapsulins and lacking portal and tail genes. The structural phylogenetic analysis of HK97-fold entities revealed bidirectional evolutionary transitions between encapsulins and HK97-fold viruses. An evolutionary mechanism responsible for such transitions was proposed based on the presence of lysogeny-associated genes in the viral entities and the structural and molecular parallels between encapsulins and procapsids, the immature state of viral capsids before genome packaging. We concluded that procapsids are akin to the common HK97-fold protein shell ancestor and might still facilitate transitions between modern viruses and encapsulins. The potential genomic and biochemical functions of HK97-fold procapsids make them an ideal candidate model for the early evolution of life and might require a revision of the role of viral capsids in the biosphere.

bioinformatics↗

bioRxiv2025

Scaling unlocks broader generation and deeper functional understanding of proteins

Generative protein language models (PLMs) are powerful tools for designing proteins purpose-built to solve problems in medicine, agriculture, and industrial processes. Recent work has trained ever larger language models, but there has been little systematic study of the optimal training distributions and the influence of model scale on the sequences generated by PLMs. We introduce the ProGen3 family of sparse generative PLMs, and we develop compute-optimal scaling laws to scale up to a 46B-parameter model pre-trained on 1.5T amino acid tokens. Pro-Gen3s pre-training data is sampled from an optimized data distribution over the Profluent Protein Atlas v1, a carefully curated dataset of 3.4B full-length proteins. We evaluate for the first time in the wet lab the influence of model scale on the sequences generated by PLMs, and we find that larger models generate viable proteins for a much wider diversity of protein families. Finally, we find both computationally and experimentally that larger models are more responsive to alignment with laboratory data, resulting in improved protein fitness prediction and sequence generation capabilities. These results indicate that larger PLMs like ProGen3-46B trained on larger, well-curated datasets are powerful foundation models that push the frontier of protein design.1

synthetic biology↗

bioRxiv2025

Engineering of CRISPR-Cas PAM recognition using deep learning of vast evolutionary data

CRISPR-Cas enzymes must recognize a protospacer-adjacent motif (PAM) to edit a genomic site, significantly limiting the range of targetable sequences in a genome. Machine learning-based protein engineering provides a powerful solution to efficiently generate Cas protein variants tailored to recognize specific PAMs. Here, we present Protein2PAM, an evolution-informed deep learning model trained on a dataset of over 45,000 CRISPR-Cas PAMs. Protein2PAM rapidly and accurately predicts PAM specificity directly from Cas proteins across Type I, II, and V CRISPR-Cas systems. Using in silico deep mutational scanning, we demonstrate that the model can identify residues critical for PAM recognition in Cas9 without utilizing structural information. As a proof of concept for protein engineering, we employ Protein2PAM to computationally evolve Nme1Cas9, generating variants with broadened PAM recognition and up to a 50-fold increase in PAM cleavage rates compared to the wild-type under in vitro conditions. This work represents the first successful application of machine learning to achieve customization of Cas enzymes for alternate PAM recognition, paving the way for personalized genome editing.

synthetic biology↗

bioRxiv2024

Adapting protein language models for structure-conditioned design

Generative models for protein design trained on experimentally determined structures have proven useful for a variety of design tasks. However, such methods are limited by the quantity and diversity of structures used for training, which represent a small, biased fraction of protein space. Here, we describe proseLM, a method for protein sequence design based on adaptation of protein language models to incorporate structural and functional context. We show that proseLM benefits from the scaling trends of underlying language models, and that the addition of non-protein context - nucleic acids, ligands, and ions - improves recovery of native residues during design by 4-5% across model scales. These improvements are most pronounced for residues that directly interface with non-protein context, which are faithfully recovered at rates >70% by the most capable proseLM models. We experimentally validated proseLM by optimizing the editing efficiency of genome editors in human cells, achieving a 50% increase in base editing activity, and by redesigning therapeutic antibodies, resulting in a PD-1 binder with 2.2 nM affinity.

bioengineering↗

bioRxiv2024

Design of highly functional genome editors by modeling the universe of CRISPR-Cas sequences

Gene editing has the potential to solve fundamental challenges in agriculture, biotechnology, and human health. CRISPR-based gene editors derived from microbes, while powerful, often show significant functional tradeoffs when ported into non-native environments, such as human cells. Artificial intelligence (AI) enabled design provides a powerful alternative with potential to bypass evolutionary constraints and generate editors with optimal properties. Here, using large language models (LLMs) trained on biological diversity at scale, we demonstrate the first successful precision editing of the human genome with a programmable gene editor designed with AI. To achieve this goal, we curated a dataset of over one million CRISPR operons through systematic mining of 26 terabases of assembled genomes and meta-genomes. We demonstrate the capacity of our models by generating 4.8x the number of protein clusters across CRISPR-Cas families found in nature and tailoring single-guide RNA sequences for Cas9-like effector proteins. Several of the generated gene editors show comparable or improved activity and specificity relative to SpCas9, the prototypical gene editing effector, while being 400 mutations away in sequence. Finally, we demonstrate an AI-generated gene editor, denoted as OpenCRISPR-1, exhibits compatibility with base editing. We release OpenCRISPR-1 publicly to facilitate broad, ethical usage across research and commercial applications.

synthetic biology↗

bioRxiv2023

You can move, but you can't hide: identification of mobile genetic elements with geNomad

Identifying and characterizing mobile genetic elements (MGEs) in sequencing data is essential for understanding their diversity, ecology, biotechnological applications, and impact on public health. Here, we introduce geNomad, a classification and annotation framework that combines information from gene content and a deep neural network to identify sequences of plasmids and viruses. geNomad uses a large dataset of marker proteins to provide functional gene annotation and taxonomic assignment of viral genomes. Using a conditional random field model, geNomad also detects proviruses integrated into host genomes with high precision. In benchmarks that included diverse MGE and chromosome sequences, geNomad significantly outperformed other tools in all evaluated clades of plasmids and viruses. Leveraging geNomads speed and scalability, we were able to process public metagenomes and metatranscriptomes, leading to the discovery of millions of new viruses and plasmids that are available through the IMG/VR and IMG/PR databases. We anticipate that geNomad will enable further advancements in MGE research, and it is available at https://portal.nersc.gov/genomad.

bioinformatics↗

bioRxiv2022

iPHoP: an integrated machine-learning framework to maximize host prediction for metagenome-assembled virus genomes

The extraordinary diversity of viruses infecting bacteria and archaea is now primarily studied through metagenomics. While metagenomes enable high-throughput exploration of the viral sequence space, metagenome-derived genomes lack key information compared to isolated viruses, in particular host association. Different computational approaches are available to predict the host(s) of uncultivated viruses based on their genome sequences, but thus far individual approaches are limited either in precision or in recall, i.e. for a number of viruses they yield erroneous predictions or no prediction at all. Here we describe iPHoP, a two-step framework that integrates multiple methods to provide host predictions for a broad range of viruses while retaining a low (<10%) false-discovery rate. Based on a large database of metagenome-derived virus genomes, we illustrate how iPHoP can provide extensive host prediction and guide further characterization of uncultivated viruses. iPHoP is available at https://bitbucket.org/srouxjgi/iphop, through a Bioconda recipe, and a Docker container.

microbiology↗

bioRxiv2022

Maast: genotyping thousands of microbial strains efficiently

Genotyping single nucleotide polymorphisms (SNPs) of intraspecific genomes is a prerequisite to performing population genetic analysis and microbial epidemiology. However, existing algorithms fail to scale for species with thousands of sequenced strains, nor do they account for the biased sampling of strains that has produced considerable redundancy in genome databases. Here we present Maast, a tool that reduces the computational burden of SNP genotyping by leveraging this genomic redundancy. Maast implements a novel algorithm to dynamically identify a minimum set of phylogenetically diverse conspecific genomes that contains the maximum number of SNPs above a user-specified allele frequency. Then it uses these genomes to construct a SNP panel for each species. A species SNP panel enables Maast to rapidly genotype thousands of strains using a hybrid of whole-genome alignment and k-mer exact matching. Maast works with both genome assemblies and unassembled sequencing reads. Compared to existing genotyping methods, Maast is more accurate and up to two orders of magnitude faster. We demonstrate Maasts utility on species with thousands of genomes by reconstructing the genetic structure of Helicobacter pylori across the globe and tracking SARS-CoV-2 diversification during the COVID-19 outbreak. Maast is a fast, reliable SNP genotyping tool that empowers population genetic meta-analysis of microbes at an unrivaled scale. Availabilitysource code of Maast is available at https://github.com/zjshi/Maast. Contactkpollard@gladstone.ucsf.edu

bioinformatics↗

bioRxiv2022

MIDAS2: Metagenomic Intra-species Diversity Analysis System

SummaryThe Metagenomic Intra-Species Diversity Analysis System (MIDAS) is a scalable metagenomic pipeline that identifies single nucleotide variants (SNVs) and gene copy number variants (CNVs) in microbial populations. Here, we present MIDAS2, which addresses the computational challenges presented by increasingly large reference genome databases, while adding functionality for building custom databases and leveraging paired-end reads to improve SNV accuracy. This fast and scalable reengineering of the MIDAS pipeline enables thousands of metagenomic samples to be efficiently genotyped. Availability and ImplementationThe source code is available at https://github.com/czbiohub/MIDAS2. The documentation is available at https://midas2.readthedocs.io/en/latest/. Supplementary InformationSupplementary data are available at Bioinformatics online.

bioinformatics↗

bioRxiv2020

Ultra-rapid metagenotyping of the human gut microbiome

Sequence variation is used to quantify population structure and identify genetic determinants of phenotypes that vary within species. In the human microbiome and other environments, single nucleotide polymorphisms (SNPs) are frequently detected by aligning metagenomic sequencing reads to catalogs of genes or genomes. But this requires high-performance computing and enough read coverage to distinguish SNPs from sequencing errors. We solved these problems by developing the GenoTyper for Prokaytotes (GT-Pro), a suite of novel methods to catalog SNPs from genomes and use exact k-mer matches to perform ultra-fast reference-based SNP calling from metagenomes. Compared to read alignment, GT-Pro is more accurate and two orders of magnitude faster. We discovered 104 million SNPs in 909 human gut species, characterized their global population structure, and tracked pathogenic strains. GT-Pro democratizes strain-level microbiome analysis by making it possible to genotype hundreds of metagenomes on a personal computer. Software availabilityGT-Pro is available at https://github.com/zjshi/gt-pro.

bioinformatics↗

bioRxiv2020

CheckV: assessing the quality of metagenome-assembled viral genomes

Over the last several years, metagenomics has enabled the assembly of millions of new viral sequences that have vastly expanded our knowledge of Earths viral diversity. However, these sequences range from small fragments to complete genomes and no tools currently exist for estimating their quality. To address this problem, we developed CheckV, which is an automated pipeline for estimating the completeness of viral genomes as well as the identification and removal of non-viral regions found on integrated proviruses. After validating the approach on mock datasets, CheckV was applied to large and diverse viral genome collections, including IMG/VR and the Global Ocean Virome, revealing that the majority of viral sequences were small fragments, with just 3.6% classified as high-quality (i.e. > 90% completeness) or complete genomes. Additionally, we found that removal of host contamination significantly improved identification of auxiliary metabolic genes and interpretation of viral-encoded functions. We expect CheckV will be broadly useful for all researchers studying and reporting viral genomes assembled from metagenomes. CheckV is freely available at: http://bitbucket.org/berkeleylab/CheckV.

bioinformatics↗

bioRxiv2019

A unified sequence catalogue of over 280,000 genomes obtained from the human gut microbiome

Comprehensive reference data is essential for accurate taxonomic and functional characterization of the human gut microbiome. Here we present the Unified Human Gastrointestinal Genome (UHGG) collection, a resource combining 286,997 genomes representing 4,644 prokaryotic species from the human gut. These genomes contain over 625 million protein sequences used to generate the Unified Human Gastrointestinal Protein (UHGP) catalogue, a collection that more than doubles the number of gut protein clusters over the Integrated Gene Catalogue. We find that a large portion of the human gut microbiome remains to be fully explored, with over 70% of the UHGG species lacking cultured representatives, and 40% of the UHGP missing meaningful functional annotations. Intra-species genomic variation analyses revealed a large reservoir of accessory genes and single-nucleotide variants, many of which were specific to individual human populations. These freely available genomic resources should greatly facilitate investigations into the human gut microbiome.

microbiology↗

bioRxiv2019

Global ecotypes in the ubiquitous marine clade SAR86

SAR86 is an abundant and ubiquitous heterotroph in the surface ocean that plays a central role in the function of marine ecosystems. We hypothesized that despite its ubiquity, different SAR86 subgroups may be endemic to specific ocean regions and functionally specialized for unique marine environments. However, the global biogeographical distributions of SAR86 genes, and the manner in which these distributions correlate with marine environments, have not been investigated. We quantified SAR86 gene content across globally-distributed metagenomic samples and modeled these gene distributions as a function of 51 environmental variables. We identified five distinct clusters of genes within the SAR86 pangenome, each with a unique geographic distribution associated with specific environmental characteristics. Gene clusters are characterized by strong taxonomic enrichment of distinct SAR86 genomes and partial assemblies, as well as differential enrichment of certain functional groups, suggesting differing functional and ecological roles of SAR86 ecotypes. We then leveraged our models and high-resolution, remote sensing-derived environmental data to predict the distributions of SAR86 gene clusters across the worlds oceans, creating global maps of SAR86 ecotype distributions. Our results reveal that SAR86 exhibits previously unknown, complex biogeography, and provide a framework for exploring geographic distributions of genetic diversity from other microbial clades.

ecology↗

bioRxiv2019

Cryptic inoviruses are pervasive in bacteria and archaea across Earth's biomes

Bacteriophages from the Inoviridae family (inoviruses) are characterized by their unique morphology, genome content, and infection cycle. To date, a relatively small number of inovirus isolates have been extensively studied, either for biotechnological applications such as phage display, or because of their impact on the toxicity of known bacterial pathogens including Vibrio cholerae and Neisseria meningitidis. Here we show that the current 56 members of the Inoviridae family represent a minute fraction of a highly diverse group of inoviruses. Using a new machine learning approach leveraging a combination of marker gene and genome features, we identified 10,295 inovirus-like genomes from microbial genomes and metagenomes. Collectively, these represent six distinct proposed inovirus families infecting nearly all bacterial phyla across virtually every ecosystem. Putative inoviruses were also detected in several archaeal genomes, suggesting that these viruses may have occasionally transferred from bacterial to archaeal hosts. Finally, we identified an expansive diversity of inovirus-encoded toxin-antitoxin and gene expression modulation systems, alongside evidence of both synergistic (CRISPR evasion) and antagonistic (superinfection exclusion) interactions with co-infecting viruses which we experimentally validated in a Pseudomonas model. Capturing this previously obscured component of the global virosphere sparks new avenues for microbial manipulation approaches and innovative biotechnological applications.

microbiology↗

bioRxiv2018

A Metagenomic Meta-Analysis Reveals Functional Signatures of Health and Disease in the Human Gut Microbiome

While recent research indicates that human health depends, in part, upon the symbiotic relationship between gut microbes and their host, the specific interactions between host and microbe that define health are poorly resolved. Metagenomic clinical studies clarify this definition by revealing gut microbial taxa and functions that stratify healthy and diseased individuals. However, the typical single-disease focus of microbiome studies limits insight into which microbiome features robustly associate with health, indicate general deviations from health, or predict specific diseases. Additionally, the focus on taxonomy may limit our understanding of how the microbiome relates to health given observations that different taxonomic members can fulfill similar functional roles. To improve our understanding of the association between the gut microbiome and health, we integrated about 2,000 gut metagenomes obtained from eight clinical studies in a statistical meta-analysis. We identify characteristics of the gut microbiome that associate generally with disease, including functional alpha-diversity, beta-diversity, and beta-dispersion. Moreover, we resolve microbiome modules that stratify diseased individuals from controls in a manner independent of study-specific effects. Many of the differentially abundant functions overlap multiple diseases suggesting a role in host health, while others are specific to a single disease and may associate with disease-specific etiologies. Our results clarify potential microbiome-mediated mechanisms of disease and reveal features of the microbiome that may be useful for the development of microbiome-based diagnostics. Ultimately, our study clarifies the definition of a healthy microbiome and how perturbations to it associate with disease.

bioinformatics↗

bioRxiv2017

A most wanted list of conserved protein families with no known domains

The number and proportion of genes with no known function are growing rapidly. To quantify this phenomenon and provide criteria for prioritizing genes for functional characterization, we developed a bioinformatics pipeline that identifies robustly defined protein families with no annotated domains, ranks these with respect to phylogenetic breadth, and identifies them in metagenomics data. We applied this approach to 271 965 protein families from the SFams database and discovered many with no functional annotation, including >118 000 families lacking any known protein domain. From these, we prioritized 6 668 conserved protein families with at least three sequences from organisms in at least two distinct classes. These Function Unknown Families (FUnkFams) are present in Tara Oceans Expedition and Human Microbiome Project metagenomes, with distributions associated with sampling environment. Our findings highlight the extent of functional novelty in sequence databases and establish an approach for creating a \"most wanted\" list of genes to characterize.

genomics↗

bioRxiv2017

Phylogeny-corrected identification of microbial gene families relevant to human gut colonization

The mechanisms by which different microbes colonize the healthy human gut versus other body sites, the gut in disease states, or other environments remain largely unknown. Identifying microbial genes influencing fitness in the gut could lead to new ways to engineer probiotics or disrupt pathogenesis. We approach this problem by measuring the statistical association between having a species having a gene and the probability that the species is present in the gut microbiome. The challenge is that closely related species tend to be jointly present or absent in the microbiome and also share many genes, only a subset of which are involved in gut adaptation. We show that this phylogenetic correlation indeed leads to many false discoveries and propose phylogenetic linear regression as a powerful solution. To apply this method across the bacterial tree of life, where most species have not been experimentally phenotyped, we use metagenomes from hundreds of people to quantify each species prevalence in and specificity for the gut microbiome. This analysis reveals thousands of genes potentially involved in adaptation to the gut across species, including many novel candidates as well as processes known to contribute to fitness of gut bacteria, such as acid tolerance in Bacteroidetes and sporulation in Firmicutes. We also find microbial genes associated with a preference for the gut over other body sites, which are significantly enriched for genes linked to fitness in an in vivo competition experiment. Finally, we identify gene families associated with higher prevalence in patients with Crohns disease, including Proteobacterial genes involved in conjugation and fimbria regulation, processes previously linked to inflammation. These gene targets may represent new avenues for modulating host colonization and disease. Our strategy of combining metagenomics with phylogenetic modeling is general and can be used to identify genes associated with adaptation to any environment.\n\nAuthor SummaryWhy do certain microbes and not others colonize our gut, and why do they differ between healthy and sick people? One explanation is the genes in their genomes. If we can find microbial genes involved in gut adaptation, we may be able to keep out pathogens and encourage the growth of beneficial microbes. One could look for genes that were present more often in prevalent microbes, and less often in rare ones.\n\nHowever, this ignores that related species are more likely to share an environment and also share many unrelated phenotypes simply because of common ancestry. To solve this problem, we used a method from ecology that accounts for phylogenetic relatedness. We first calculated gut prevalence for thousands of species using a compendium of shotgun sequencing data, then tested for genes associated with prevalence, adjusting for phylogenetic relationships. We found genes that are associated with overall gut prevalence, with a preference for the gut over other body sites, and with the gut in Crohns disease versus health. Many of these findings have biological plausibility based on existing literature. We also showed agreement with the results of a previously published high-throughput screen of bacterial gene knockouts in mice. These results, and this type of analysis, may eventually lead to new strategies for maintaining gut health.

bioinformatics↗