This repository contains configuration, GIS, analysis files used for modeling the prevalence of malaria (P. falciparum) in Rwanda. These configurations are primarily intended to study 1) the development of antimalarial resistance in the parasite, and 2) possible interventions that the National Malaria Control Programme can implement.

The origin repository for the simulation can be found at maciekboni/PSU-CIDD-Malaria-Simulation, although the studies in this repository are intended to be run against 4.x series versions that can be found under rjzupkoii/PSU-CIDD-Malaria-Simulation.

Analysis - This directory primary contains scripts related to model calibration, data processing, and preliminary analysis of data.

GIS - GIS files for Rwanda, as prepared for the simulation. Note that two versions of the PfPR_2-10 files are included. One based upon the original version of the Malaria Atlas project projections (rwa_pfpr2to10_v2017.asc) and one based upon the updated 2020 version (rwa_pfpr_v2020.asc). The rwa_pfpr2to10.asc is the canonical file for generating the beta values.

Plotting - This directory primarily contains scripts that related to generating plots; either for analysis, or for publication in manuscripts.

Replicates - Contains the job files used to run replicates on the compute cluster, primarily configured for SLURM.

Scripts - Scripts used for model calibration, analysis, and other exploratory purposes.

Studies - YAML configuration files used to define the Rwanda model, along with other configurations defining various interventions.

The Rwanda model requires ~8GB of RAM during model execution and thus can be run on reasonably performant workstations. However, a minimum of 50 valid replicates are needed for analysis and 100 valid replicates is recommended. As such, most simulations were performed on the Pennsylvania State University’s Institute for Computational and Data Sciences’ Roar /supercomputer using the configurations present in Studies.

The following dependencies need to be installed for all of the scripts included in this repository to run:

• numpy : https://pypi.org/project/numpy/
• scipy : https://pypi.org/project/scipy/

Some scripts written in Matlab are present, although most work is done in Python. The preference is for future work to be done exclusively in Python.