COVID-19 in Latin America: Contrasting phylodynamic inference with epidemiological surveillance. (Molecular epidemiology of COVID-19 in Latin America)

  1. Diana M. Rojas-Gallardo
  2. Sandra C. Garzón-Castaño
  3. Natalia Millán
  4. Erika V. Jiménez-Posada
  5. Marlen Martínez-Gutiérrez
  6. Julian Ruiz-Saenz
  7. Jaime A. Cardona-Ospina

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

Background

SARS-CoV-2 revealed important gaps in infectious disease surveillance. Molecular epidemiology can help monitoring and adapting traditional surveillance to surpass those limitations. This work aims to contrast data driven from traditional surveillance with parameters inferred from molecular epidemiology in Latin America (LATAM)

Methods

We obtained epidemiological data up to 4 th June, 2020. We estimated Effective Reproductive Number (Re) and epidemic curves using maximum likelihood (ML). SARS-CoV-2 genomes were obtained from GISAID up to June 4 th 2020. We aligned sequences, generated a ML phylogenetic tree, and ran a coalescent model Birth Death SIR. Phylodynamic analysis was performed for inferring Re, number of infections and date of introduction.

Findings

A total of 1,144,077 cases were reported up to 4 th June 2020. Countries with the largest cumulative cases were Chile, Peru and Panama. We found at least 18 different lineages circulating, with a predominance of B.1 and B.1.1. We inferred an underestimation of the daily incident cases. When contrasting observed and inferred Re, we did not find statistically significant differences except for Chile and Mexico. Temporal analysis of the introduction of SARS-CoV-2 suggested a detection lag of at least 21 days.

Interpretation

Our results support that epidemiological and genomic surveillance are two complementary approaches. Even with a low number of genomes proper estimations of Re could be performed. We suggest that countries, especially developing countries, should consider to add genomic surveillance to their systems for monitoring and adapting epidemiological control of SARS-CoV-2.

Article activity feed

  1. Version published to 10.1101/2020.05.23.20111443v2 on medRxiv
  2. ScreenIT

    SciScore for 10.1101/2020.05.23.20111443: (What is this?)

    Please note, not all rigor criteria are appropriate for all manuscripts.

    Table 1: Rigor

    NIH rigor criteria are not applicable to paper type.

    Table 2: Resources

    Software and Algorithms
    SentencesResources
    Sequences were aligned using MAFFT software v7.450 and visually inspected and edited using Bioedit v7.0.5.3.
    MAFFT
    suggested: (MAFFT, RRID:SCR_011811)
    Bioedit
    suggested: (BioEdit, RRID:SCR_007361)
    To assess for viral recombination, we employed the pairwise homoplasy index (PHI) test using SplitsTree v4.15.1.
    SplitsTree
    suggested: (SplitsTree, RRID:SCR_014734)
    Nucleotide substitution model for the alignment was inferred according to the corrected Akaike information criterion (AICc) in jModelTest v2.1.10.
    jModelTest
    suggested: (jModelTest, RRID:SCR_015244)
    The ML tree was inferred using PhyML software V3.0 (DOI: 10.1093/sysbio/syq010) with an approximate likelihood-ratio test (aLRT) approach.
    PhyML
    suggested: (PhyML, RRID:SCR_014629)
    Re evolution through time was estimated using the skyspline R package 0.1, and SIR trajectories were calculated using the phydynR R package 0.2.017 using an exponential growth model using birth rates, death rates and effective population number (Ne) estimated by BEAST for each country.
    BEAST
    suggested: (BEAST, RRID:SCR_010228)

    Results from OddPub: We did not detect open data. We also did not detect open code. Researchers are encouraged to share open data when possible (see Nature blog).

    Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:
    Nonetheless, new available data can help to surpass this limitation, particularly for Brazil. According to Candido et al,20 when a data set of 427 new sequences were added to the GISAID available sequences up to 30 April, a strong spatial representativity of the country was found. Therefore, a subsequent analysis of this data could provide insights in to the real magnitude of the pandemic in this country. On the other hand, although our estimations of the percentage of population infected for Argentina, Brazil, Chile and Colombia significantly differ from modelled seroprevalence estimations for those countries,21 our estimations of the percentage of undiagnosed infections are similar to those showed by seroprevalence studies. According to our results, at least the 96% of the infections in the analyzed countries (excluding Brazil) are not diagnosed, that agrees with serological community based studies in Hubei, China that suggests that 97% of infections might have gone undiagnosed during the epidemic, 22 and in Geneva, Switzerland that showed that for every confirmed case there were 11.6 infections in the community.23 Regarding transmissibility, the Re was in good agreement for every country, except for Chile and Mexico, and is consistent with previous estimations for Latin American countries24-27. This has two important implications: First, the good agreement at this respect validates previous transmission dynamics of viral shedding estimated by He, X, et al,14 suggesting tha...

    Results from TrialIdentifier: No clinical trial numbers were referenced.

    Results from Barzooka: We did not find any issues relating to the usage of bar graphs.

    Results from JetFighter: We did not find any issues relating to colormaps.

    Results from rtransparent:
    • Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
    • Thank you for including a funding statement. Authors are encouraged to include this statement when submitting to a journal.
    • No protocol registration statement was detected.

    About SciScore

    SciScore is an automated tool that is designed to assist expert reviewers by finding and presenting formulaic information scattered throughout a paper in a standard, easy to digest format. SciScore checks for the presence and correctness of RRIDs (research resource identifiers), and for rigor criteria such as sex and investigator blinding. For details on the theoretical underpinning of rigor criteria and the tools shown here, including references cited, please follow this link.

  3. Version published to 10.1101/2020.05.23.20111443v1 on medRxiv