Fig. 1: Phylodynamic analysis of the SARS-CoV-2 Omicron BA.1 invasion in England.

a,b, A summary of estimates from a simultaneous estimation of sequence and discrete (geographic) trait data with a GLM extension of the discrete-trait model and two epochs (with 25 December 2021 as the transition time), showing the effect size estimates for a subset of GLM predictors considered² for both epochs (a) and the mean Markov jump estimates between 256 lower-tier local authorities (LTLAs) for the expansion phase epoch, ordered in a clockwise fashion first by being part of Greater London (blue) or not (yellow) and then by population size (b). c,d, A spread.gl³⁰ visualization of a RRW model fit to latitudes and longitudes (randomly drawn from within the LTLA of sampling) for the same large transmission lineage. Part of the maximum clade credibility tree is projected up to 13 December 2021 (c) and 25 December 2021 (d), respectively. The arcs represent dispersal events with a light- and dark-blue color from origin to destination, respectively. e, Comparison of doubling-time estimates based on an exponential growth coalescent model applied to about 1,000 genomes sampled during the expansion phase of Omicron BA.1 and Alpha (B.1.1.7) in England based on 10,000 posterior samples. f, Summary of estimates of effective population size (Ne) under a nonparameteric skygrid coalescent model²⁸ and estimates of the effective reproduction number (Re) under an episodic birth–death sampling model⁹, based on 6,000 and 3,000 posterior samples, respectively. The yellow-shaded area in the Re plot represents the time period during which ‘Plan B’ measures were implemented in the UK. Box plots in a and e show the median (middle quartile) as a thick line, the box represents the upper and lower quartiles, and the whiskers indicate the 95% highest posterior density interval, whereas f shows the 95% credible intervals in blue and the median in black.