Millions of new viral sequences have been identified from metagenomes, but the quality and completeness of these sequences vary considerably. Here we present CheckV, an automated pipeline for identifying closed viral genomes, estimating the completeness of genome fragments and removing flanking host regions from integrated proviruses. CheckV estimates completeness by comparing sequences with a large database of complete viral genomes, including 76,262 identified from a systematic search of publicly available metagenomes, metatranscriptomes and metaviromes. After validation on mock datasets and comparison to existing methods, we applied CheckV to large and diverse collections of metagenome-assembled viral sequences, including IMG/VR and the Global Ocean Virome. This revealed 44,652 high-quality viral genomes (that is, >90% complete), although the vast majority of sequences were small fragments, which highlights the challenge of assembling viral genomes from short-read metagenomes. Additionally, we found that removal of host contamination substantially improved the accurate identification of auxiliary metabolic genes and interpretation of viral-encoded functions.

Identifying and characterizing mobile genetic elements 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 dataset of more than 200,000 marker protein profiles 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, geNomad achieved high classification performance for diverse plasmids and viruses (Matthews correlation coefficient of 77.8% and 95.3%, respectively), substantially outperforming other tools. Leveraging geNomad’s speed and scalability, we processed over 2.7 trillion base pairs of sequencing data, leading to the discovery of millions of viruses and plasmids that are available through the IMG/VR and IMG/PR databases. geNomad is available at https://portal.nersc.gov/genomad.

Viruses are widely recognized as critical members of all microbiomes. Metagenomics enables large-scale exploration of the global virosphere, progressively revealing the extensive genomic diversity of viruses on Earth and highlighting the myriad of ways by which viruses impact biological processes. IMG/VR provides access to the largest collection of viral sequences obtained from (meta)genomes, along with functional annotation and rich metadata. A web interface enables users to efficiently browse and search viruses based on genome features and/or sequence similarity. Here, we present the fourth version of IMG/VR, composed of >15 million virus genomes and genome fragments, a ≈6-fold increase in size compared to the previous version. These clustered into 8.7 million viral operational taxonomic units, including 231 408 with at least one high-quality representative. Viral sequences in IMG/VR are now systematically identified from genomes, metagenomes, and metatranscriptomes using a new detection approach (geNomad), and IMG standard annotation are complemented with genome quality estimation using CheckV, taxonomic classification reflecting the latest taxonomic standards, and microbial host taxonomy prediction. IMG/VR v4 is available at https://img.jgi.doe.gov/vr, and the underlying data are available to download at https://genome.jgi.doe.gov/portal/IMG_VR.

Metagenomes encode an enormous diversity of proteins, reflecting a multiplicity of functions and activities^1,2. 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.

Historically neglected by microbial ecologists, soil viruses are now thought to be critical to global biogeochemical cycles. However, our understanding of their global distribution, activities and interactions with the soil microbiome remains limited. Here we present the Global Soil Virus Atlas, a comprehensive dataset compiled from 2,953 previously sequenced soil metagenomes and composed of 616,935 uncultivated viral genomes and 38,508 unique viral operational taxonomic units. Rarefaction curves from the Global Soil Virus Atlas indicate that most soil viral diversity remains unexplored, further underscored by high spatial turnover and low rates of shared viral operational taxonomic units across samples. By examining genes associated with biogeochemical functions, we also demonstrate the viral potential to impact soil carbon and nutrient cycling. This study represents an extensive characterization of soil viral diversity and provides a foundation for developing testable hypotheses regarding the role of the virosphere in the soil microbiome and global biogeochemistry.

The advent of high-throughput sequencing technologies made it possible to obtain large volumes of genetic information, quickly and inexpensively. Thus, many efforts are devoted to unveiling the biological roles of genomic elements, being the distinction between protein-coding and long non-coding RNAs one of the most important tasks. We describe RNAsamba, a tool to predict the coding potential of RNA molecules from sequence information using a neural network-based that models both the whole sequence and the ORF to identify patterns that distinguish coding from non-coding transcripts. We evaluated RNAsamba’s classification performance using transcripts coming from humans and several other model organisms and show that it recurrently outperforms other state-of-the-art methods. Our results also show that RNAsamba can identify coding signals in partial-length ORFs and UTR sequences, evidencing that its algorithm is not dependent on complete transcript sequences. Furthermore, RNAsamba can also predict small ORFs, traditionally identified with ribosome profiling experiments. We believe that RNAsamba will enable faster and more accurate biological findings from genomic data of species that are being sequenced for the first time. A user-friendly web interface, the documentation containing instructions for local installation and usage, and the source code of RNAsamba can be found at https://rnasamba.lge.ibi.unicamp.br/.