This repo contains code and data to support Finch et al "Climate variation and serotype competition drive dengue outbreak dynamics in Singapore"(in press). This study quantifies the role of climate variation and serotype competition in shaping dengue risk in Sinapore using over 20 years of weekly case data, and integrates these findings into an early warning system framework able to predict dengue outbreaks up to 2 months ahead. Model fitting is performed in INLA. Climate and dengue case data are available in the data folder. Note that data on dengue serotype frequencies shown in Figure 1 are not included in this repo as they are not publicly available.
To reproduce the analysis, open the .Rproj file. The following scripts read in the data and run the analysis:
The following scripts contain functions needed for analysis. These do not need to be run seperately as they are sourced in the main workflow scripts.
Model outputs are included in the output folder and manuscript figures can be generated by knitting reports/manuscript-figures.Rmd.
This analysis was performed using R 4.1.1 and INLA_21.11.22. This analysis was performed on a Mac OSX system with 32 Gb RAM and 10 cores. Please note the final step is computationally intensive and takes several hours to run under these specifications for a given forecast horizon.