Review 1: "Estimating the increased transmissibility of the B.1.1.7 strain over previously circulating strains in England using fractions of GISAID sequences and the distribution of serial intervals"

Christine Tedijanto

doi:doi:10.1162/2e3983f5.e8ce246d

RR:C19 Evidence Scale rating by reviewer:

Reliable. The main study claims are generally justified by its methods and data. The results and conclusions are likely to be similar to the hypothetical ideal study. There are some minor caveats or limitations, but they would/do not change the major claims of the study. The study provides sufficient strength of evidence on its own that its main claims should be considered actionable, with some room for future revision.

***************************************

Review:

In this article, the authors propose a likelihood-based method to estimate the selective advantage, defined as a proportional increase in the reproduction number, of a new strain over existing strains. The method uses strain frequencies over time and the serial interval distribution. Applying this method to GISAID sequences from September 2020 to mid-February 2021, the authors estimate the selective advantage of the Alpha variant to be 34.4% (26.6-45.9% when considering uncertainty in the serial interval) in England. The paper is clearly and concisely written.

For the most part, the assumptions made by the authors appear reasonable. One exception is the assumption that the number of infections at previous time steps is fairly similar (Equation 7). In Figure 1, it is clear that the number of sequences (and presumably the number of infections) may differ widely over a 20-day period. It may be worth investigating how modifying this assumption would impact results.

The estimated selective advantage of the Alpha variant in this paper was lower than previously published estimates (in some cases considerably so), and this paper could be strengthened by exploring other factors that may contribute to this discrepancy. The authors address this in the discussion but focus on differing serial intervals. Differing datasets, time frames, and/or geographic limits of analysis may also play a role. For example, the assumption of England as a single well-mixed population may attenuate the result, since the method uses overall strain frequency and does not account for heterogeneity in emergence times across different populations. Discussion of such factors would be helpful for framing the method and results of this paper within the context of existing literature.

The authors propose a method for estimating the relative transmissibility of new strains that will be useful as variants continue to emerge. Minor revisions would enhance this work prior to publication.