A speed-fidelity trade-off determines the mutation rate and virulence of an RNA virus

William Fitzsimmons; Robert J. Woods; John T. McCrone; Andrew Woodman; Jamie J. Arnold; Madhumita Yennawar; Richard Evans; Craig E. Cameron; Adam S. Lauring

4B, Table S2 ), but there was a clear delay in the 3D G64S group (p=0.0239, p<0.001, p<0.001 for 225 10 5 , 10 6 , 10 7 pfu inocula, respectively). This lag persisted even when we increased the number 226 of rare, but undetectable, variants in the inoculum by infecting with doses 20-fold higher than the 227 LD 50 of 3D G64S (Table S3 ). Both viruses spread to the CNS and replicated to high titers after 228 intravenous inoculation, although WT titers in the brain and spinal cord were marginally higher 229 at 5 days post infection ( Figure 4C , p=0.0012 for brain and p=0.0221 for spinal cord, Mann 230 Whitney U test). Here too, WT mice generally developed severe morbidity more rapidly than 231 those infected with 3D G64S . We characterized the mutations present in the CNS populations of 10 were rare, none were shared among mice, and there was an excess of synonymous or 234 noncoding variants relative to nonsynonymous ones. These data are consistent with random 235 sampling of the infecting population as opposed to positive selection.