Genotyping-By-Sequencing (GBS) Detects Genetic Structure and Confirms Behavioral QTL in Tame and Aggressive Foxes (Vulpes vulpes)

Abstract

The silver fox (Vulpes vulpes) offers a novel model for studying the genetics of social behavior and animal domestication. Selection of foxes, separately, for tame and for aggressive behavior has yielded two strains with markedly different, genetically determined, behavioral phenotypes. Tame strain foxes are eager to establish human contact while foxes from the aggressive strain are aggressive and difficult to handle. These strains have been maintained as separate outbred lines for over 40 generations but their genetic structure has not been previously investigated. We applied a genotyping-by-sequencing (GBS) approach to provide insights into the genetic composition of these fox populations. Sequence analysis of EcoT22I genomic libraries of tame and aggressive foxes identified 48,294 high quality SNPs. Population structure analysis revealed genetic divergence between the two strains and more diversity in the aggressive strain than in the tame one. Significant differences in allele frequency between the strains were identified for 68 SNPs. Three of these SNPs were located on fox chromosome 14 within an interval of a previously identified behavioral QTL, further supporting the importance of this region for behavior. The GBS SNP data confirmed that significant genetic diversity has been preserved in both fox populations despite many years of selective breeding. Analysis of SNP allele frequencies in the two populations identified several regions of genetic divergence between the tame and aggressive foxes, some of which may represent targets of selection for behavior. The GBS protocol used in this study significantly expanded genomic resources for the fox, and can be adapted for SNP discovery and genotyping in other canid species.

Comparative genomic evidence for self-domestication inHomo sapiens

“…FAM172A, selected for in dogs, cattle, and AMH, may perhaps be worthy of note given its position on chromosome 5 neighboring NR2F1, which plays a role in regulating neural crest specifier genes and has undergone selection in AMHs 32,33 ; The functionally related nuclear receptor NR2F2, involved in regulating embryonic stem-cell differentiation 34 , and implicated in neural crest development, has been selected for in the domesticated fox 35 .…”

“…Likewise, more information about the molecular basis of the famous domestication experiment with the silver fox (Vulpes vulpes) 104 should prove valuable. As a matter of fact, the only study focusing on the genetic divergence between foxes that were selected for tame and aggressive behavior 35 reveals an intriguing overlap between a set of SNPs significantly differentiating the two fox strains and among the genes noted in the silver fox data 35 and a number of genes considered to fall within non-coding human accelerated regions (ncHARs) 105 . The most significant among these genes are those implicated in neural tube and forebrain development, and are listed by Racimo (2016) 28 as having been selected from within a single region on chromosome 12 in the AMH Eurasian branch.…”

“…In the end, the lists of selected genes for these species (compared to their wild counterparts) were too small to draw any firm conclusion. To help us understand domestication-related changes better, we made use of the comparison of two lines of rats (Rattus norvegicus) selected for tame and aggressive behaviour to identify genetic loci that differ between the lines 112 , the comparison of gene expression levels in the brains of domesticated and wild animals offered in 90 , genomic signatures of domestication in neurogenetic genes in Drosophila melanogaster, where neurogenetic genes have been claimed to be associated with signs of positive selection 91 , and the genetic divergence between foxes (Vulpes vulpes) that were selected for tame and aggressive behavior 35 .…”

Chicken domestication : Effects of tameness on brain gene expression and DNA methylation

“…And drift may by chance increase or decrease the allelic frequencies, which complicates things when trying to differentiate between effects of selection or drift [171]. In the silver fox populations, genetic diversity remains despite many generations of selective and separated breeding, and some regions diverge and may be targets for selection [172]. Long term selection on complex traits is the result of involvement of potentially hundreds of genes, and selection even in small populations will affect a large number of these, such as the case with growth in chickens [139].…”

Adapting Genotyping-by-Sequencing for Rice F2 Populations

“…Performing GBS on an F2 population incurs some specific difficulties since 50 % of all SNP sites are expected to be in a heterozygous state. This demands higher read coverage to accurately call genotypes, since correctly calling a heterozygous allele requires the presence of reads from both allele states (Johnson et al 2015; Hyma et al 2015). Some existing GBS pipelines and imputation algorithms deal with that problem by omitting heterozygous calls.…”