Search

Sequencing, assembly, and annotation of environmental virome samples is challenging. Methodological biases and differences in species abundance result in fragmentary read coverage; sequence reconstruction is further complicated by the mosaic nature of viral genomes. In this paper, we focus on biocomputational aspects of virome analysis, emphasizing latent pitfalls in sequence annotation. Using simulated viromes that mimic environmental data challenges we assessed the performance of five assemblers (CLC-Workbench, IDBA-UD, SPAdes, RayMeta, ABySS). Individual analyses of relevant scaffold length fractions revealed shortcomings of some programs in reconstruction of viral genomes with excessive read coverage (IDBA-UD, RayMeta), and in accurate assembly of scaffolds ⩾50 kb (SPAdes, RayMeta, ABySS). The CLC-Workbench assembler performed best in terms of genome recovery (including highly covered genomes) and correct reconstruction of large scaffolds; and was used to assemble a virome from a copper rich site in the Namib Desert. We found that scaffold network analysis and cluster-specific read reassembly improved reconstruction of sequences with excessive read coverage, and that strict data filtering for non-viral sequences prior to downstream analyses was essential. In this study we describe novel viral genomes identified in the Namib Desert copper site virome. Taxonomic affiliations of diverse proteins in the dataset and phylogenetic analyses of circovirus-like proteins indicated links to the marine habitat. Considering additional evidence from this dataset we hypothesize that viruses may have been carried from the Atlantic Ocean into the Namib Desert by fog and wind, highlighting the impact of the extended environment on an investigated niche in metagenome studies. [ABSTRACT FROM AUTHOR]

Viral communities of two different salt pans located in the Namib Desert, Hosabes and Eisfeld, were investigated using a combination of multiple displacement amplification of metaviromic DNA and deep sequencing, and provided comprehensive sequence data on both ssDNA and dsDNA viral community structures. Read and contig annotations through online pipelines showed that the salt pans harbored largely unknown viral communities. Through network analysis, we were able to assign a large portion of the unknown reads to a diverse group of ssDNA viruses. Contigs belonging to the subfamily Gokushovirinae were common in both environmental datasets. Analysis of haloarchaeal virus contigs revealed the presence of three contigs distantly related with His1, indicating a possible new lineage of salterproviruses in the Hosabes playa. Based on viral richness and read mapping analyses, the salt pan metaviromes were novel and most closely related to each other while showing a low degree of overlap with other environmental viromes. [ABSTRACT FROM AUTHOR]

Recent years have seen the classification and reclassification of many viruses related to the model enterobacterial phage P2. Here, we report the identification of a prophage (Smhb1) that infects Salinivibrio kushneri BNH isolated from a Namib Desert salt pan (playa). Analysis of the genome revealed that it showed the greatest similarity to P2-like phages that infect Vibrio species and showed no relation to any of the previously described Salinivibrio-infecting phages. Despite being distantly related to these Vibrio infecting phages and sharing the same modular gene arrangement as seen in most P2-like viruses, the nucleotide identity to its closest relatives suggest that, for now, Smhb1 is the lone member of the Peduovirus genus Playavirus. Although host range testing was not extensive and no secondary host could be identified for Smhb1, genomic evidence suggests that the phage is capable of infecting other Salinivibrio species, including Salinivibrio proteolyticus DV isolated from the same playa. Taken together, the analysis presented here demonstrates how adaptable the P2 phage model can be. [ABSTRACT FROM AUTHOR]