Disclaimer: These analyses must be interpreted with caution and estimated times of events may not be reliable
Phylogeographic analyses are only as good as the data we input. For these analyses we have assumed that the temporal scaling of the phylogenetic tree reliable and that the genetic diversity of the sequences is representative of the Omicron epidemic in Norway. Most of the early samples of Omicron in Norway came from a super-spreading event at Aker Brygge. Thus, there is a substantial overrepresentation of the Aker Brygge outbreak in the genetic data, which could influence the results.
Geographical mapping of omicron lineages
The analyses take a dated phylogeny exported from NextStrain as a starting point for quantifying imports and local transmission of SARS-CoV-2 Omicron variant in Norway.The current set of sequences includes 2727 Norwegian genome sequences, and 1300 from the “Rest of the world (RoW)” Each isolate is assigned to one of two binary categories: Norway or RoW. The output is based on maximum likelihood ancestral character estimation performed on the dated phylogeny and presented in Fig.1.
Test
Fig. 1. Omicron phylogeny colored by binary geographic mappings (Norway, rest of the world (RoW).
The large group in Norway in top of the plot represents the large outbreak at Aker Brygge and downstream transmission from this super-spreading event (SSE). The other groups in Norway in different parts of the tree represents separate introductions from the SSE.
Lineage size distributions
Below is a visualization of the size of each transmission lineage detected in Norway
Fig. 2. Omicron transmission lineage sizes in Norway. G denotes the number assigned to each lineage, S denotes the size of the lineage and TMRCA denotes the age of the lineage. The area of the squares represents the size of the lineages.
Quantifying introductions and local transmission
Based on the estimated locations of the lineages in Fig. 1 we can can derive the number of importation events, and local transmission events using the methods from LineageHomology. LineageHomology assumes that each lineage observed in Norway starts with one importation event, and that each subsequent branching event in the lineage represents local transmission events.
Table 1. Estimated number of infections in Norway resulting from import and local transmission based on the phylogeographic analyses in Figure 1.
| 2.5% | 50% | 97.5% | |
|---|---|---|---|
| Import | 348.4750 | 366.0000 | 379.5250 |
| Local transmission | 2347.4750 | 2361.0000 | 2378.5250 |
| Import / Total | 0.1278 | 0.1342 | 0.1392 |
Next, we estimate the fraction of introductions relative to local transmission over time. In these and subsequent analyses it is important to keep detection lag into account (see https://science.sciencemag.org/content/371/6530/708). There will always be a delay between the introduction and detection, which varies from case to case. In the figures below, the last four weeks prior to the time of sampling of the most recent included isolate is shaded to indicate that estimates in this period is highly uncertain.
Fig. 4. Relative quantification of infection source (new introduction / import vs local transmission). Grey shading indicates that inferences in this period (last 4 weeks) are uncertain due to detection lag.
The same output is visualized as absolute numbers in Fig. 5 below. It should be noted that absolute estimates will be significantly shaped by sampling density (i.e. if we only obtain 20% samples of the true cases in a transmission lineage, the true size would in reality be five times larger).
Fig. 5. Absolute quantification of imports and local transmission events. Grey shading indicates that inferences in this period (last 4 weeks) are uncertain due to detection lag.
Growth and size of individual lineages
Perhaps the most interesting aspect of quantifying introductions, is to investigate the fate of the introduced viruses. Mandatory testing at the border and/or mandatory quaranteens etc will obviously have a bearing on the fate of new introductions from abroad. The plot below visualizes the number of infections in Norway resulting from each identified virus import event over time, stratified by inferred TMRCA of each introduced lineage.
Fig. 6. Transmission lineage sizes as a function of time of introduction. “TMRCA” = time of most recent common ancestor. Grey shading indicates that inferences in this period (last 4 weeks) are uncertain due to detection lag. NB. we have excluded the largest lineage that is visible in
