Related Experiment Video
Updated: May 14, 2026
Protocol for Plasmodium falciparum Infections in Mosquitoes and Infection Phenotype Determination
Published on: July 4, 2007
Assessing the impact of temporal changes in transmission on Plasmodium falciparum strains in Asembo, western Kenya
Gary Vestal1, Zhiyong Zhou2, Sheila Sergent3
1Department of Computer Science, Emory University, Atlanta, USA. gary.kenneth.vestal@emory.edu.
Background:
Genomic surveillance of malaria parasites offers important insights into the impact of interventions on transmission reduction and changes in pathogen populations over time, especially in low-transmission areas. However, such surveillance faces challenges in high-transmission regions. Detecting temporal changes in transmission in high-transmission settings requires analytical methods tailored to high-diversity parasite populations that can differentiate between superinfection (infection through multiple mosquito bites, each bearing an unrelated strain) and co-transmission (infection through a single mosquito bite bearing more than one strain).
Methods:
This study applied a previously developed novel Next Generation Sequencing (NGS) 24-SNP barcode assay for genotyping smear-positive samples obtained from a 2017 cross-sectional survey in the Asembo area, western Kenya, building on previous work on samples collected in the surveys conducted in 1996, 2001, 2007, and 2012. Algorithms StrainRecon and STIM were used to identify parasite strains within a sample and measure multiplicity of infection (MOI). Population genetic metrics of FST (Fixation Index), strain-relatedness by IBD (Identity by Descent), Hs (Modified Heterozygosity) and Ne (Effective Population Size) were evaluated using the same 24-SNP data. This study further explored a novel slope metric of the relationship between within-host strain relatedness and MOI to infer superinfection and co-transmission. Temporal changes in the above metrics were assessed.
Results:
There was no significant differentiation in FST, Hs, Ne, and strain-relatedness at the population level over time. In contrast, the average MOI significantly decreased from 4.32 in 1996 to 3.34 in 2012, although it increased to 3.49 in 2017. Insecticide-treated bednet distribution campaigns from 1997 to 2017 did track these temporal changes in MOI. Additionally, the value of strain relatedness within-host (IBD) was inversely correlated with MOI (number of strains), and the change in the inverse relationship (within-host slopes) over time was verified by two different correlation analysis and modelling. The temporal trends in this within-host slope metric suggested that transmission dynamics shifted towards co-transmission from 2001 to 2012, and then returned to similar levels of superinfection in 2017 as in 1996.
Conclusion:
The within-host MOI, IBD-based strain-relatedness, and their mathematical relationship (slope) provide useful metrics for understanding the transmission dynamics in our study. Notably, this study presents the first simple slope-based method using 24-SNP barcodes to distinguish superinfection from co-transmission in a high transmission area, warranting further evaluation of the novel tool in other high transmission settings.
