Unifying viral genetics and human transportation data to predict the global transmission dynamics of human influenza H3N2.
Lemey P, Rambaut A, Bedford T, Faria N, Bielejec F, Baele G, Russell CA, Smith DJ, Pybus OG, Brockmann D, Suchard MA. · 2014
Read Unifying viral genetics and human transportation data to predict the global transmission dynamics of human influenza H3N2. by Lemey P, Rambaut A, Bedford T, Faria N, Bielejec F, Baele G, Russell CA, Smith DJ, Pybus OG, Brockmann D, Suchard MA. (2014) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc3930559.txt.
Introduction
Philippe Lemey and colleagues combine virus family trees with human travel data to predict how flu spreads worldwide, in PLoS Pathogens in 2014. Knowing how people move is central to forecasting influenza, yet case counts alone cannot show which routes matter most. Gene sequences from viruses carry a record of where lineages have been, but earlier analyses rarely used mobility data directly.
Working with the H3N2 flu subtype, the team built a model that reconstructs the virus's migration history while testing candidate drivers, including air travel patterns, at the same time. The framework can weigh how well each factor explains the observed spread and can forecast spatial patterns for influenza A more broadly. It turns two separate data streams into one test of global transmission routes.
Catalog record
| CREATOR | Lemey P, Rambaut A, Bedford T, Faria N, Bielejec F, Baele G, Russell CA, Smith DJ, Pybus OG, Brockmann D, Suchard MA. |
|---|---|
| YEAR | 2014 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 56 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Unifying Viral Genetics and Human Transportation Data to Predict the Global Transmission Dynamics of Human Influenza H3N2 By Philippe Lemey, Andrew Rambaut, Trevor Bedford, Nuno Faria, Filip Bielejec, Guy Baele, Colin A Russell, Derek J Smith, Oliver G Pybus, Dirk Brockmann, Marc A Suchard, Neil M Ferguson Published in: PLoS Pathogens DOI: 10.1371/journal.ppat.1003932 ABSTRACT Abstract Information on global human movement patterns is central to spatial epidemiological models used to predict the behavior of influenza and other infectious diseases. Yet it remains difficult to test which modes of dispersal drive pathogen spread at various geographic scales using standard epidemiological data alone. Evolutionary analyses of pathogen genome sequences increasingly provide insights into the spatial dynamics of influenza viruses, but to date they have largely neglected the wealth of information on human mobility, mainly because no statistical framework exists within which viral gene sequences and empirical data on host movement can be combined. Here, we address this problem by applying a phylogeographic approach to elucidate the global spread of human influenza subtype H3N2 and assess its ability to predict the spatial spread of human influenza A viruses worldwide. Using a framework that estimates the migration history of human influenza while simultaneously testing and quantifying a range of potential predictive variables of spatial spread, we show that the global dynamics of influenza H3N2 are driven by air passenger flows, whereas at more local scales spread is also determined by processes that correlate with geographic distance. Our analyses further confirm a central role for mainland China and Southeast Asia in maintaining a source population for global influenza diversity. By comparing model output with the known pandemic expansion of H1N1 during 2009, we demonstrate that predictions of influenza spatial spread are most accurate when data on human mobility and viral evolution are integrated. In conclusion, the global dynamics of influenza viruses are best explained by combining human mobility data with the spatial information inherent in sampled viral genomes. The integrated approach introduced here offers great potential for epidemiological surveillance through phylogeographic reconstructions and for improving predictive models of disease control. INTRODUCTION The emergence and worldwide dispersal of novel human pathogens is increasingly challenging global public health [1]. Notable recent examples include novel influenza strains, severe acute respiratory syndrome (SARS) virus and Methicillin-resistant Staphylococcus aureus, which all exploit today's complex and voluminous transport networks to rapidly disseminate in a globalized world. In the context of human infectious diseases, the worldwide air transportation network is by far the best studied system of global mobility [2]. Air travel likely drives the global circulation of seasonal influenza A (H3N2) viruses [3], and may explain seasonal dynamics in the absence of locally-persistent strains between epidemic seasons. Retrospective modeling of the ‘Hong Kong flu’ H3N2 pandemic in 1968 indicates that the virus spread through a global network of cities interconnected by air travel [4]. Numerous modeling and simulation studies have subsequently explored the potential influence of air travel on influenza virus spread, e.g. [5]–[8], but few have attempted to verify such models against underlying empirical data on human movement patterns [9]. Two studies on the timing and rate of seasonal influenza transmission across the United States of America (USA) highlight the difficulty of using standard epidemiological data to disentangle the relative contributions of different human transportation systems to influenza spread. Using weekly time series of excess mortality due to pneumonia and influenza (P&I), Viboud et al. [9] demonstrated that the patterns of timing and incidence of outbreaks across the continental USA are most strongly associated with rates of movement of people to and from their workplaces, and to a lesser extent with the distance between locations and various measures of domestic transportation. In contrast, Brownstein et al. [10] concluded that the rate of inter-regional spread and timing of influenza in the USA, as measured using weekly P&I mortality statistics, is predicted by domestic airline travel volume in November. These discordant findings generated significant debate [11], especially in the context of a potential pandemic of pathogenic influenza [12], which would require rapid decisions to be made on the implementation of travel restrictions. As a historical record of epidemic spread, viral genetic sequence data may offer a valuable source of information for the empirical verification of epidemiological models. Several studies have demonstrated their utility and power, for example by revealing the genetic dynamics of influenza A H3N2 seasonality [13] and the spatial patterns of global H3N2 circulation [3], [14]. More generally, it is recognized that the genetic diversity of rapidly evolving viruses like influenza should be analysed in a framework that unifies evolutionary and ecological dynamics [15]. Current attempts to reconstruct viral spread through time and space from genetic data, however, typically fit parameter-rich models to sparse spatial data and result in phylogeographic patterns that are difficult to relate directly to underlying ecological processes [16]. Together with potential sampling bias, this complicates phylogeographic tasks, such as the characterization of source-sink dynamics in seasonal influenza. It is therefore unsurprising that different studies on the global circulation of H3N2 are sometimes inconsistent [3], [14], [17], despite the importance of such work for influenza surveillance and vaccine strain selection. Here we use a model-based approach to explicitly tests spatial epidemiological hypotheses by integrating empirical data on human movement patterns with viral genetic data. This framework enables us to measure the relative contribution of different predictive variables to viral spatial spread. We apply this approach to seasonal H3N2 dynamics and use it to identify key drivers of the global dissemination of influenza viruses. Analysis of different sampling schemes, including one that represents the community structure in global air transportation, provides consistent support for air travel governing the spatial dynamics of seasonal H3N2 infections. Using epidemiological simulations, we further demonstrate that estimates resulting from the merger of human air travel and H3N2 influenza genetics best capture the observed global expansion of pandemic H1N1 influenza in 2009. METHODS Sequence data We complemented a previously collected hemagglutinin sequence data set, comprising 1,441 sequences sampled globally from 2002 to 2007 [3], with publicly available sequences sampled within the same time interval. The allocation of the sequence data into 15 and 26 geographic regions as well as into 14 air communities is described in detail in Supporting information Text S1. Air transportation data and modularity maximization The worldwide air transportation network is defined by a passenger flux matrix that quantifies the number of passengers traveling between each pair of airports. We use a dataset provided by OAG (Official Airline Guide) Ltd. (http://www.oag.com), containing 4,092 airports and the number of seats on scheduled commercial flights between pairs of airports during the years 2004–2006. We take the number of seats on scheduled commercial flights from airport i to j to be proportional to the number of passengers traveling. To identify air transportation communities, we approximate a maximal-modularity subdivision of the 1,227-largest-airport network by employing a recently described stochastic Monte-Carlo approach [18]. Modularity provides a measure of how well the connectivity of a network is described by partitioning its nodes into non-overlapping groups; for a definition we refer to [19]. For any given partition, modularity will be high if connectivity within groups is high and connectivity among groups is low. For large networks, a variety of