More Like this

1. Discovery, diversity and distribution of functional dark matter through global metagenomics.

   Pavlopoulos, G.A. ( October 2023 , Nature)

   Metagenomes encode an enormous diversity of proteins, reflecting a multiplicity of functions and activities1,2. Exploration of this vast sequence space has been limited to a comparative analysis against reference microbial genomes and protein families derived from those genomes. Here, to examine the scale of yet untapped functional diversity beyond what is currently possible through the lens of reference genomes, we develop a computational approach to generate reference-free protein families from the sequence space in metagenomes. We analyse 26,931 metagenomes and identify 1.17 billion protein sequences longer than 35 amino acids with no similarity to any sequences from 102,491 reference genomes or the Pfam database3. Using massively parallel graph-based clustering, we group these proteins into 106,198 novel sequence clusters with more than 100 members, doubling the number of protein families obtained from the reference genomes clustered using the same approach. We annotate these families on the basis of their taxonomic, habitat, geographical and gene neighbourhood distributions and, where sufficient sequence diversity is available, predict protein three-dimensional models, revealing novel structures. Overall, our results uncover an enormously diverse functional space, highlighting the importance of further exploring the microbial functional dark matter. 

   more » « less
2. Complementary Metagenomic Approaches Improve Reconstruction of Microbial Diversity in a Forest Soil

   https://doi.org/10.1128/mSystems.00768-19  

   Alteio, L. V. ; Schulz, F. ; Seshadri, R. ; Varghese, N. ; Rodriguez-Reillo, W. ; Ryan, E. ; Goudeau, D. ; Eichorst, S. A. ; Malmstrom, R. R. ; Bowers, R. M. ; et al ( April 2020 , mSystems)

   Jansson, Janet K. (Ed.)

   ABSTRACT Soil ecosystems harbor diverse microorganisms and yet remain only partially characterized as neither single-cell sequencing nor whole-community sequencing offers a complete picture of these complex communities. Thus, the genetic and metabolic potential of this “uncultivated majority” remains underexplored. To address these challenges, we applied a pooled-cell-sorting-based mini-metagenomics approach and compared the results to bulk metagenomics. Informatic binning of these data produced 200 mini-metagenome assembled genomes (sorted-MAGs) and 29 bulk metagenome assembled genomes (MAGs). The sorted and bulk MAGs increased the known phylogenetic diversity of soil taxa by 7.2% with respect to the Joint Genome Institute IMG/M database and showed clade-specific sequence recruitment patterns across diverse terrestrial soil metagenomes. Additionally, sorted-MAGs expanded the rare biosphere not captured through MAGs from bulk sequences, exemplified through phylogenetic and functional analyses of members of the phylum Bacteroidetes . Analysis of 67 Bacteroidetes sorted-MAGs showed conserved patterns of carbon metabolism across four clades. These results indicate that mini-metagenomics enables genome-resolved investigation of predicted metabolism and demonstrates the utility of combining metagenomics methods to tap into the diversity of heterogeneous microbial assemblages. IMPORTANCE Microbial ecologists have historically used cultivation-based approaches as well as amplicon sequencing and shotgun metagenomics to characterize microbial diversity in soil. However, challenges persist in the study of microbial diversity, including the recalcitrance of the majority of microorganisms to laboratory cultivation and limited sequence assembly from highly complex samples. The uncultivated majority thus remains a reservoir of untapped genetic diversity. To address some of the challenges associated with bulk metagenomics as well as low throughput of single-cell genomics, we applied flow cytometry-enabled mini-metagenomics to capture expanded microbial diversity from forest soil and compare it to soil bulk metagenomics. Our resulting data from this pooled-cell sorting approach combined with bulk metagenomics revealed increased phylogenetic diversity through novel soil taxa and rare biosphere members. In-depth analysis of genomes within the highly represented Bacteroidetes phylum provided insights into conserved and clade-specific patterns of carbon metabolism. 

   more » « less
3. Network-Based Prediction of Novel CRISPR-Associated Genes in Metagenomes

   https://doi.org/10.1128/mSystems.00752-19  

   Weissman, Jake L. ; Johnson, Philip L. ( February 2020 , mSystems)

   Chia, Nicholas (Ed.)

   ABSTRACT A diversity of clustered regularly interspaced short palindromic repeat (CRISPR)-Cas systems provide adaptive immunity to bacteria and archaea through recording “memories” of past viral infections. Recently, many novel CRISPR-associated proteins have been discovered via computational studies, but those studies relied on biased and incomplete databases of assembled genomes. We avoided these biases and applied a network theory approach to search for novel CRISPR-associated genes by leveraging subtle ecological cooccurrence patterns identified from environmental metagenomes. We validated our method using existing annotations and discovered 32 novel CRISPR-associated gene families. These genes span a range of putative functions, with many potentially regulating the response to infection. IMPORTANCE Every branch on the tree of life, including microbial life, faces the threat of viral pathogens. Over the course of billions of years of coevolution, prokaryotes have evolved a great diversity of strategies to defend against viral infections. One of these is the CRISPR adaptive immune system, which allows microbes to “remember” past infections in order to better fight them in the future. There has been much interest among molecular biologists in CRISPR immunity because this system can be repurposed as a tool for precise genome editing. Recently, a number of comparative genomics approaches have been used to detect novel CRISPR-associated genes in databases of genomes with great success, potentially leading to the development of new genome-editing tools. Here, we developed novel methods to search for these distinct classes of genes directly in environmental samples (“metagenomes”), thus capturing a more complete picture of the natural diversity of CRISPR-associated genes. 

   more » « less
4. Deep learning of a bacterial and archaeal universal language of life enables transfer learning and illuminates microbial dark matter

   https://doi.org/10.1038/s41467-022-30070-8  

   Hoarfrost, A. ; Aptekmann, A. ; Farfañuk, G. ; Bromberg, Y. ( May 2022 , Nature Communications)

   The majority of microbial genomes have yet to be cultured, and most proteins identified in microbial genomes or environmental sequences cannot be functionally annotated. As a result, current computational approaches to describe microbial systems rely on incomplete reference databases that cannot adequately capture the functional diversity of the microbial tree of life, limiting our ability to model high-level features of biological sequences. Here we present LookingGlass, a deep learning model encoding contextually-aware, functionally and evolutionarily relevant representations of short DNA reads, that distinguishes reads of disparate function, homology, and environmental origin. We demonstrate the ability of LookingGlass to be fine-tuned via transfer learning to perform a range of diverse tasks: to identify novel oxidoreductases, to predict enzyme optimal temperature, and to recognize the reading frames of DNA sequence fragments. LookingGlass enables functionally relevant representations of otherwise unknown and unannotated sequences, shedding light on the microbial dark matter that dominates life on Earth.

    

   more » « less
5. The contribution of historical processes to contemporary extinction risk in placental mammals

   https://doi.org/10.1126/science.abn5856  

   Wilder, Aryn P. ; Supple, Megan A. ; Subramanian, Ayshwarya ; Mudide, Anish ; Swofford, Ross ; Serres-Armero, Aitor ; Steiner, Cynthia ; Koepfli, Klaus-Peter ; Genereux, Diane P. ; Karlsson, Elinor K. ; et al ( April 2023 , Science)

   INTRODUCTION The Anthropocene is marked by an accelerated loss of biodiversity, widespread population declines, and a global conservation crisis. Given limited resources for conservation intervention, an approach is needed to identify threatened species from among the thousands lacking adequate information for status assessments. Such prioritization for intervention could come from genome sequence data, as genomes contain information about demography, diversity, fitness, and adaptive potential. However, the relevance of genomic data for identifying at-risk species is uncertain, in part because genetic variation may reflect past events and life histories better than contemporary conservation status. RATIONALE The Zoonomia multispecies alignment presents an opportunity to systematically compare neutral and functional genomic diversity and their relationships to contemporary extinction risk across a large sample of diverse mammalian taxa. We surveyed 240 species spanning from the “Least Concern” to “Critically Endangered” categories, as published in the International Union for Conservation of Nature’s Red List of Threatened Species. Using a single genome for each species, we estimated historical effective population sizes ( N e ) and distributions of genome-wide heterozygosity. To estimate genetic load, we identified substitutions relative to reconstructed ancestral sequences, assuming that mutations at evolutionarily conserved sites and in protein-coding sequences, especially in genes essential for viability in mice, are predominantly deleterious. We examined relationships between the conservation status of species and metrics of heterozygosity, demography, and genetic load and used these data to train and test models to distinguish threatened from nonthreatened species. RESULTS Species with smaller historical N e are more likely to be categorized as at risk of extinction, suggesting that demography, even from periods more than 10,000 years in the past, may be informative of contemporary resilience. Species with smaller historical N e also carry proportionally higher burdens of weakly and moderately deleterious alleles, consistent with theoretical expectations of the long-term accumulation and fixation of genetic load under strong genetic drift. We found weak support for a causative link between fixed drift load and extinction risk; however, other types of genetic load not captured in our data, such as rare, highly deleterious alleles, may also play a role. Although ecological (e.g., physiological, life-history, and behavioral) variables were the best predictors of extinction risk, genomic variables nonrandomly distinguished threatened from nonthreatened species in regression and machine learning models. These results suggest that information encoded within even a single genome can provide a risk assessment in the absence of adequate ecological or population census data. CONCLUSION Our analysis highlights the potential for genomic data to rapidly and inexpensively gauge extinction risk by leveraging relationships between contemporary conservation status and genetic variation shaped by the long-term demographic history of species. As more resequencing data and additional reference genomes become available, estimates of genetic load, estimates of recent demographic history, and accuracy of predictive models will improve. We therefore echo calls for including genomic information in assessments of the conservation status of species. Genomic information can help predict extinction risk in diverse mammalian species. Across 240 mammals, species with smaller historical N e had lower genetic diversity, higher genetic load, and were more likely to be threatened with extinction. Genomic data were used to train models that predict whether a species is threatened, which can be valuable for assessing extinction risk in species lacking ecological or census data. [Animal silhouettes are from PhyloPic] 

   more » « less