> The wide geographic ranges of R. pusillus and R. affinis and the fact that two of the closest known relatives of SARS-CoV-2, RpYN06 and RaTG13, have been samp

> The wide geographic ranges of R. pusillus and R. affinis and the fact that two of the closest known relatives of SARS-CoV-2, RpYN06 and RaTG13, have been sampled in these species flags them as prime suspects for the source of the SARS-CoV-2’s progenitor in China. Additionally, these two bat species are found in shared roosts with R. sinicus and R. ferrumequinum in Yunnan and with R. sinicus in Guangxi[30], providing opportunities for host switches, co-infections and thus recombination between the sarbecoviruses that these bat species carry. R. pusillusand R. affinisal so link more regions of China with bat species such as R. shameli, R. malayanusand R. acuminatus which are only found in Southeast Asia and southwest of China(Figure 4B). Latinne et al. (2020)[31] recently published a large-scale sampling expedition of coronaviruses across bats in China. Although only short RdRp fragments were sequenced, reconstructing the phylogeny for the novel viruses reveals a cluster of seven identical Sarbecovirus sequences sampled from R. affinis within the nCoV clade, that cluster close to SARS-CoV-2 (Figure S2). Still, the fact that viruses in the Yunnan clade (consisting of RmYN02, RpYN06 and PrC31) were sampled from three different Rhinolophus species supports the hypothesis that these viruses readily infect multiple different horseshoe bat species with overlapping geographical ranges. > Based on the analysis of the Sarbecovirus and host data presented here, we propose that horseshoe bat population sampling should focus on the known ranges of probable bat hosts. Specifically, samples should be collected in roosting environments spread across China with care being taken both to avoid a further spillover or reverse zoonosis and to protect the bat species[30]. Sampling strategies will also need to consider the distinct subspecies of Rhinolophus as the delineators of genetically meaningful host populations for coronaviruses. For example, there are two R. affinis sub-species on mainland China: himalayanus and macrurus[32]. Future sampling should also encompass a range of indigenous mammals other than bats that we now know can be infected by coronaviruses. Although highly endangered, it is possible that Chinese pangolins, given their susceptibility to infection and their geographical range across southern China[33], might be the “missing” intermediate host of the SARS-CoV-2 proximal ancestor[1]. > Conclusion > The currently available data, although sparse, illustrates a complex reticulate history involving the lineage of Sarbecoviruses SARS-CoV-2 emerged from, governed by co-circulation of related coronaviruses, over at least the last 100 years, across the bat populations in southern China, and into Southeast Asia with multiple recombination events imprinted on the genomes of these viruses. Considering the high frequency of recombination, it is expected that selection could preferentially favour exchanges of specific genome regions, in line with our detection of hotspots near the Spike gene (Figure 1B,C). Our analysis further illustrates the importance of accounting for recombination rather than using whole-genome pairwise similarity to determine the shared evolutionary history of these viruses. This is exemplified by RaTG13 which is often described as the “on average” closest Sarbecovirus to SARS-CoV-2 despite not being the closest virus once recombination history is accounted for in the othernCoV Sarbecoviruses (Figure 1A, 3A). > The evidence of recombination events between Sarbecoviruses sampled in different geographical regions and from different bat hosts, indicates recent extensive movement of the viruses between different regions and species (and presumably sub-species too) as a result of the different bat populations that carry them coming into contact with one another. Although very few nCoV viruses are known to be hosted by mammals other than horseshoe bats, the recombination patterns detected within the nCoV genomes imply the existence of one or a few primary reservoir hosts with a geographical range spanning Thailand from the Southwest and Zhejiang to the East, a distribution that is consistent with specific Chinese horseshoe bats being the primary reservoir hosts. Having presented evidence in support of R. affinis and R. pusillus’s potential significance as the reservoir species,it should be noted that at least 20 different Rhinolophus species are distributed across China (four being endemic to China), many of which have not yet been found hosting nCoVs. The generalist nature of Sarbecoviruses also means multiple wild or farmed animals (e.g., minks)[1] could have facilitated transmission of SARS-CoV-2 from bats to humans. > The risk of future emergence of a new SARS-CoV-2 like nCoV strain in humans is too high to restrict sampling strategies. Beyond the relatively rare detection of SARS-like antibodies in people from rural communities in China[34,35],SARS related coronaviruses have not, to our knowledge, seeded outbreaks in humans before. This indicates that there is limited human exposure to these viruses, suggesting ecological “barriers”to their emergence[36]. One possible recent disruption that caused widespread and unusual movements of animals in China and may have increased the permeability of this barrier, was the dramatic shortage of pork products in 2019[37] attributable to an African swine fever virus (ASFV) outbreak impacting 100s of millions of pigs in China[38]. Such a major disruption in the food supply chain will have potentially brought humans into increased contact with Sarbecovirus infected animals as i) exotic meats replaced pork, ii) animals from rural locations were brought to city markets and/or iii) by meat from infected animals being transported in cold chain processes. Further facilitating animal contacts,increased human encroachment into rural areas as a result of new and faster travel networks around and between metropolitan areas have also likely increased opportunities for Sarbecoviruses to spillover into humans.Given the reality of frequent human-animal contact, routine characterisation of respiratory infections would seem a sensible precaution to prevent future emergence of Sarbecoviruses. > The key and most urgent questions relating to the prevention of another emergence, is thus not how did SARS-CoV-2 get from Yunnan to Hubei, but rather which bat or other animal species are harbouring nCoV Sarbecoviruses and what are the risks of a future spillover? There is undoubtedly a virus highly related to SARS-CoV-2 still present somewhere most probably in a bat species in South China or towards Yunnan in the southwest. The best we can do is maximize the probability that future sampling efforts will uncover that host species or sub-species.