Abstract
Knowledge of how malaria infections spread locally is important both for the design of targeted interventions aiming to interrupt malaria transmission and the design of trials to assess the interventions. A previous analysis of 1602 genotyped Plasmodium falciparum parasites in Kilifi, Kenya collected over 12 years found an interaction between time and geographic distance: the mean number of single nucleotide polymorphism (SNP) differences was lower for pairs of infections which were both a shorter time interval and shorter geographic distance apart. We determine whether the empiric pattern could be reproduced by a simple model, and what mean geographic distances between parent and offspring infections and hypotheses about genotype-specific immunity or a limit on the number of infections would be consistent with the data. We developed an individual-based stochastic simulation model of households, people and infections. We parameterized the model for the total number of infections, and population and household density observed in Kilifi. The acquisition of new infections, mutation, recombination, geographic location and clearance were included. We fit the model to the observed numbers of SNP differences between pairs of parasite genotypes. The patterns observed in the empiric data could be reproduced. Although we cannot rule out genotype-specific immunity or a limit on the number of infections per individual, they are not necessary to account for the observed patterns. The mean geographic distance between parent and offspring malaria infections for the base model was 0.4 km (95% CI 0.24, 1.20), for a distribution with 58% of distances shorter than the mean. Very short mean distances did not fit well, but mixtures of distributions were also consistent with the data. For a pathogen which undergoes meiosis in a setting with moderate transmission and a low coverage of infections, analytic methods are limited but an individual-based model can be used with genotyping data to estimate parameter values and investigate hypotheses about underlying processes.
Highlights
The spatial and temporal patterns of malaria infections at the community level are made up of movements of individual mosquitoes, human hosts and the parasites inside them
Knowing the dispersal distance from breeding sites does not lead to a calculation from which we can predict the mixing of parasite genotypes
We aimed to determine which processes and parameter values were consistent with the observed patterns of genetic differences over time and distance in Kilifi, where the observed numbers of single nucleotide polymorphism (SNP) different were fewer for parasite pairs with both a shorter time interval and a shorter geographic distance apart[3]
Summary
The spatial and temporal patterns of malaria infections at the community level are made up of movements of individual mosquitoes, human hosts and the parasites inside them. On a sub-national scale, a study from school surveys in western Kenya, an area of high malaria transmission, found no spatial structure to genetic relatedness over geographic distance in a well mixing parasite population[4]. The number of SNPs different increased with time, but saturated as distance between parasites increased[3] While these studies provide information on the distribution of malaria parasite genotypes over space and time with qualitative inferences on parasite mixing, the underlying parameters and processes leading to these patterns have not been well established and the geographic distances between parent and offspring infections remain unclear. Knowing the dispersal distance from breeding sites does not lead to a calculation from which we can predict the mixing of parasite genotypes
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
