SamuelBrand1 / KenyaCoV

This repo contains the source code KenyaCoV, a simulation package for forecasting SARS-CoV-2 transmission in Kenya.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

KenyaCoV - Forecasting for SARS-CoV-2 transmission in Kenya.

This version of the KenyaCoV model is now depreciated. The goal of this modelling exercise was to forecast the potential size of the SARS-CoV-2 epidemic in Kenya before wide-spread transmission and under a wide range of scenarios. Since, the pandemic has arrived in Kenya we have gone forwards with modelling which has been calibrated to Kenyan data.

Additionally, we have the following projects ongoing:

  • KenyaCoVMultistrains: Mechanistic modelling of the spread of SARS-CoV-2 variants into and around Kenya.
  • KenyaCoVaccines: More detailed and realistic modelling of disease burden with vaccination rollout forecasting.

The legacy code in this repository has been updated from the original pre-print version to include extra realism in the possible disease pathways of infected individuals. The modelling results of the pre-print can be recovered by the new version of KenyaCoV by setting the average period of pre-symptomatic transmission to 0, and combining mild and mild-then-severe cases. Therefore, we are not maintaining a legacy version of the KenyaCoV model used in the pre-print.

This repository contains source code for transmission modelling, forecasting, and visualisation for the SARS-CoV-2/COVID-19 epidemic in Kenya.

The main model is based on the wider population groupings of Wesolowski et al. (2012), with explicit age structure. Within population group age mixing is given by the Prem et al estimate for Kenya. A pre-print giving modelling forecasts for Kenya, and describing the underlying mathematical structure of KenyaCoV can now be found on Medrxiv (https://www.medrxiv.org/content/10.1101/2020.04.09.20059865v1).

About

This repo contains the source code KenyaCoV, a simulation package for forecasting SARS-CoV-2 transmission in Kenya.

License:MIT License


Languages

Language:Jupyter Notebook 81.9%Language:HTML 10.6%Language:Julia 7.2%Language:MATLAB 0.3%