Abstract
BackgroundAssociation studies are a promising way to uncover the genetic basis of complex traits in wild populations. Data on population stratification, linkage disequilibrium and distribution of variant effect-sizes for different trait-types are required to predict study success but are lacking for most taxa. We quantified and investigated the impacts of these key variables in a large-scale association study of a strongly selected trait of medical importance: pyrethroid resistance in the African malaria vector Anopheles gambiae.Methodology/Principal FindingsWe genotyped ≈1500 resistance-phenotyped wild mosquitoes from Ghana and Cameroon using a 1536-SNP array enriched for candidate insecticide resistance gene SNPs. Three factors greatly impacted study power. (1) Population stratification, which was attributable to co-occurrence of molecular forms (M and S), and cryptic within-form stratification necessitating both a partitioned analysis and genomic control. (2) All SNPs of substantial effect (odds ratio, OR>2) were rare (minor allele frequency, MAF<0.05). (3) Linkage disequilibrium (LD) was very low throughout most of the genome. Nevertheless, locally high LD, consistent with a recent selective sweep, and uniformly high ORs in each subsample facilitated significant direct and indirect detection of the known insecticide target site mutation kdr L1014F (OR≈6; P<10−6), but with resistance level modified by local haplotypic background.ConclusionPrimarily as a result of very low LD in wild A. Gambiae, LD-based association mapping is challenging, but is feasible at least for major effect variants, especially where LD is enhanced by selective sweeps. Such variants will be of greatest importance for predictive diagnostic screening.
Highlights
Anopheles gambiae is the most important vector of malaria, which endangers almost half the world’s population and causes nearly a million deaths annually, most in children under five [1]
Our results demonstrate that molecular forms are differentiated at a high and consistent enough level throughout the genome to cause massive inflation of test results, which could not be corrected by statistical genomic control
Our data illustrate that association mapping is difficult but feasible for wild A. gambiae populations, at least for major effect variants and especially, where Linkage disequilibrium (LD) is enhanced by selective sweeps
Summary
Anopheles gambiae is the most important vector of malaria, which endangers almost half the world’s population and causes nearly a million deaths annually, most in children under five [1]. Several types of mechanisms might produce resistant phenotypes in Anopheles mosquitoes, but available evidence suggests that the most important are alterations to the target site of the insecticide and metabolic resistance via altered expression of detoxification genes [5]. In A. gambiae, DNA polymorphisms associated with metabolic resistance in field samples have yet to be discovered, and the relative importance of target site resistance and metabolic detoxification mechanisms in natural populations is unclear [13]. Comprehensive candidate gene screens of natural mosquito populations represent a logical and promising step to identify a broader spectrum of genetic variants important in insecticide resistance and could pave the way for subsequent genomewide association (GWA) studies. We quantified and investigated the impacts of these key variables in a large-scale association study of a strongly selected trait of medical importance: pyrethroid resistance in the African malaria vector Anopheles gambiae
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
