Predicting the global mammalian viral sharing network using phylogeography

Gregory F Albery; Evan A Eskew; Noam Ross; Kevin J Olival

Most emerging human viruses originate in wild mammals, so understanding the drivers of 27 interspecific viral transmission in these taxa is an important public health research priority 1,2 . 28 Despite a rapidly expanding knowledge base, the mammalian viruses known to science 29 remain taxonomically biased and limited in scope, likely comprising less than 1% of the 30 complete mammalian virome 3,4 . Furthermore, host range is inadequately characterized even 31 for the best-studied viruses 5-7 . To help prioritise viral discovery efforts and zoonotic disease 32 surveillance in wildlife, studies have revealed high (zoonotic) parasite diversity in certain 33 host taxa, such as rodents and bats 5, 8 , and/or linked parasite diversity with host phenotypic 34 traits such as reproductive output 9,10 . Viral diversity has also been associated with host 35 macroecological traits, including geographic range size 11 and sympatry with other mammals 5 . 36 The rationale for investigating viral diversity is that species with more viruses will generate 37 more opportunities for viral transmission to other species, including humans. However, in 38 order to infect a new host species, a virus must transmit, invade, and potentially replicate 39 within the novel host 12 . Each of these processes becomes less likely if the two hosts differ 40 more in terms of their geographic range, behaviour, and/or biochemistry (i.e., cellular 41 receptors allowing viral attachment and invasion) 12, 13 . Consequently, the probability that a 42 pair of hosts will share a virus is shaped both by the species' underlying viral diversity and by 43 species interactions represented by pairwise measures such as spatial overlap, phylogenetic 44 relatedness, and ecological similarity [14] [15] [16] . 45