Timing the SARS-CoV-2 index case in Hubei province

Abstract

Understanding when severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) emerged is critical to evaluating our current approach to monitoring novel zoonotic pathogens and understanding the failure of early containment and mitigation efforts for COVID-19. We used a coalescent framework to combine retrospective molecular clock inference with forward epidemiological simulations to determine how long SARS-CoV-2 could have circulated before the time of the most recent common ancestor of all sequenced SARS-CoV-2 genomes. Our results define the period between mid-October and mid-November 2019 as the plausible interval when the first case of SARS-CoV-2 emerged in Hubei province, China. By characterizing the likely dynamics of the virus before it was discovered, we show that more than two-thirds of SARS-CoV-2-like zoonotic events would be self-limited, dying out without igniting a pandemic. Our findings highlight the shortcomings of zoonosis surveillance approaches for detecting highly contagious pathogens with moderate mortality rates.

Fig. 1. Posterior distribution for the tMRCA of 583 sampled SARS-CoV-2 genomes circulating in China between December 2019 and April 2020.

Inference was performed by using a strict molecular clock and a Bayesian Skyline coalescent prior. The shaded area denotes 95% HPD. The long-dashed line represents 17 November 2019, and the short-dashed line represents 1 December 2019.

Fig. 2. Hypothetical coalescent scenarios depicting how the time between index case infection and time of stable coalescence can vary on the basis of stochastic extinction events of basal viral lineages.

Coalescence can occur within or contemporaneously with the index case (upper left) or, in cases infected later in the course of the epidemic, with one (upper right) or more (lower left) basal lineages going extinct. In extreme cases, the epidemic can persist at low levels for a long time before stable coalescence (lower right).

Fig. 3. Forward simulations estimating the timing of the index case in Hubei province.

(A) Days between index case infection and stable coalescence in forward compartmental epidemic simulations (n = 1000). (B and C) Days between index case infection and stable coalescence after rejection sampling (B) and posterior distribution for date of index case infection (C), conditioned on an ascertained case by 17 November 2019, which is denoted by a long-dashed line. (D and E) Epidemic simulation, conditioned on an ascertained case by 1 December 2019, which is denoted by a short-dashed line. (F to H) Two-phase epidemic (F) days between index case carrying less-fit variant and adaptation (n = 2000), (G) days between adaptation and stable coalescence (n = 1000), and (H) posterior distribution for date of index case infection, conditioned on an ascertained case by 17 November 2019. (I to K) Two-phase epidemic conditioned on an ascertained case by 1 December 2019. Gray dashed lines indicate median estimates.

Fig. 4. Epidemic growth in compartmental simulations.

(A) Estimated total number of people infected in late 2019. Dark purple shading represents central 50% HPD, intermediate purple shading represents central 95% HPD, and light purple represents central 99% HPD. (B) Estimated total number of people infected in late 2019 for a two-phase epidemic. (C) Number of people infected over time in a sample of epidemic simulations that established (purple; n = 30) and went extinct (gray; n = 70). The y axis transitions to log scale once 10 people are infected at any given time. The lower panel shows the proportion of simulations that still have at least 1 infected individual over time (persisting epidemics in purple; extinct epidemics in gray). (D) Sample (n = 10) of two-phase epidemic simulations transitioning from less-fit phase 1 (blue) to more-fit phase 2 (purple). Each line represents a single simulation and its transition over time. The lower panel shows the average proportion of phase 1 to phase 2infected individuals over time.