Evolutionary origins of the SARS-CoV-2 sarbecovirus lineage responsible for the COVID-19 pandemic

Maciej F Boni; Philippe Lemey; Xiaowei Jiang; Tommy Tsan-Yuk Lam; Blair Perry; Todd Castoe; Andrew Rambaut; David L Robertson

. CC-BY-NC 4.0 International license author/funder. It is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/2020.03.30.015008 doi: bioRxiv preprint Figure 1 . Top: Breakpoints identified by 3SEQ shown by the percentage of sequences (out of 68) that support a particular breakpoint position. Note that breakpoints can be shared between sequences if they are descendants of the same recombination events. The pink, green, and orange bars show breakpoint-free regions (BFRs), with Region A (nt showing two trimmed segments to yield Region A' (nt 13291-14932, 15405-17162, 18009-19628) . Region B spans nucleotides 3625-9150, and region C spans 9261-11795. The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/2020.03.30.015008 doi: bioRxiv preprint 0.04 Y u n n a n 2 0 1 6 [ Y N 2 0 1 8 A ] Y u n n a n 2 0 1 2 [ R f4092] Y u n n a n 2 0 1 3 [ R s4231] To avoid artefacts due to recombination, we focused on the non-recombining regions NRR1, NRR2, and the recombination-masked alignment NRA3 for inferring time-measured evolutionary histories. Visual exploration using TempEst indicates there is no evidence for temporal signal in these data sets ( Figure S1 ). This is not surprising for diverse viral populations with relatively deep evolutionary histories. In such cases, even moderate rate variation among long deep phylogenetic branches will significantly impact expected root-to-tip divergences over a sampling time range that represents only a small fraction of the evolutionary history (Trova et al., 2015) . However, formal testing using marginal likelihood estimation (Duchene et al., 2019) does not reject the absence of temporal signal in all three data sets (Table S1) , albeit without strong support in favor of temporal signal (log Bayes factor support of 3, 10, and 3 for NRR1, NRR2, and NRA3 respectively).