Ecogenomics and potential biogeochemical impacts of globally abundant ocean 1 viruses 2

Ocean microbes drive global-scale biogeochemical cycling 1 , but do so under constraints imposed by 47 viruses on community composition, metabolic activity, and evolutionary trajectories 2,3 . Due to 48 sampling and cultivation challenges, genome-level viral diversity remains poorly described and grossly 49 understudied in nature such that <1% of observed surface ocean viruses are ‘known’ 4 . Here we 50 assemble complete genomes and large genomic fragments from both surface and deep ocean viruses 51 sampled during the Tara Oceans and Malaspina research expeditions 5,6 and analyze the resulting 52 Global Ocean Viromes (GOV) dataset to present a global map of abundant, double stranded DNA 53 (dsDNA) viruses complete with genomic and ecological contexts. A total of 15,222 epi- and 54 mesopelagic viral populations were identified that comprised 867 viral clusters (VCs, approximately 55 genus-level groups 7,8 ). This roughly triples the number of ocean viral populations 4 , doubles candidate 56 bacterial and archaeal virus genera 8 , and near-completely samples epipelagic communities at both the 57 population and VC level. Thirty-eight of the 867 VCs were locally or globally abundant and together 58 accounted for nearly half of the viral populations in any GOV sample. While two thirds of them 59 represent newly described viruses that lacked any cultivated representative, most could be 60 computationally linked to dominant, ecologically relevant microbial hosts. Moreover, we identified 243 61 viral-encoded auxiliary metabolic genes (AMGs), only 95 of which were known. Deeper analyses of 62 four of these AMGs ( dsr C, sox YZ, P-II and amo C) revealed that abundant viruses may directly 63 manipulate sulfur and nitrogen cycling throughout the epipelagic ocean. This viral catalog and 64 functional analyses provide a critically-needed foundation to begin meaningfully integrating viruses 65 into ecosystem models as key players in nutrient cycling and trophic networks.