A genotyping array for the globally invasive vector mosquito, Aedes albopictus

Abstract

Abstract Background Although whole genome sequencing (WGS) is the preferred genotyping method for most genomic analyses, limitations are often experienced when studying genomes characterized by a high percentage of repetitive elements, high linkage, and recombination deserts. The Asian tiger mosquito (Aedes albopictus), for example, has a genome comprised of up to 72% repetitive elements and therefore we set out to develop a SNP (Single Nucleotide Polymorphism) chip to be more cost-effective. Ae. albopictus is an invasive species originating from Southeast Asia that has recently spread around the world and is vector for many human diseases. Genome research considered a vital approach to develop biological control methods and study population demography of this pest species thus development of an accessible genotyping platform has important implications for public health. Methods We designed a SNP chip for Ae. albopictus (Aealbo chip) based on approximately 2.7 million SNPs identified using 819 WGS data from worldwide samples. We validated the chip using laboratory single-pair crosses, comparing technical replicates, and comparing genotypes of samples genotyped by WGS and the SNP chip. We then used the chip for a population genomic analysis of 237 samples from 28 sites in the native range to evaluate its usefulness to describe patterns of genomic variation and trace the origins of invasions. Results Probes on the Aealbo chip target 175,296 SNPs in coding and noncoding regions across all three chromosomes, with a density of 102 SNPs per 1Mb window, and at least one SNP in each of the 17,461 protein-coding genes. Overall, 70% of the probes captured the genetic variation. Segregation analysis found 98% of the SNPs followed expectations of single-copy Mendelian genes. Comparisons with WGS indicated that sites with genotype disagreements were mostly heterozygotes at loci with WGS read depth < 20 while there was near complete agreement with WGS read depths > 20 indicating the chip more accurately detects heterozygotes than low coverage WGS. Sample sizes did not affect accuracy of the SNP chip genotype calls. Ancestry analyses identified four to five genetic clusters in the native range with various levels of admixture. Conclusions The Aealbo chip is highly accurate, is concordant with genotypes from WGS with high sequence coverage and may be more accurate than low coverage WGS.