Skip to main content
ARS Home » Plains Area » Miles City, Montana » Livestock and Range Research Laboratory » Research » Publications at this Location » Publication #327545
Title: Use of observed genomic information to infer linkage disequilibrium between markers and QTLs

Author
item Hay, El Hamidi
item REKAYA, ROMDHANE - University Of Georgia

Submitted to: Agricultural Sciences
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/29/2018
Publication Date: 11/26/2018
Citation: Hay, E.A., Rekaya, R. 2018. Use of observed genomic information to infer linkage disequilibrium between markers and QTLs. Agricultural Sciences. 9(11):4170-4178. https://doi.org/10.4236/as.2018.911102.
DOI: https://doi.org/10.4236/as.2018.911102

Interpretive Summary: Genomic selection is a new approach of improving economically important traits in the field of animal agriculture. This form of selection relies on high number of genetic markers that are in linkage disequilibrium (LD) with quantitative trait loci (QTLs) affecting traits of interest. Higher linkage disequilibrium results in a higher probability of genetic markers and QTLs to get inherited together. However, LD changes across breeds which make genomic selection ineffective in a multi-breed genomic selection scenario. In this study, the change of LD across two breeds was predicted using the genetic markers information. The results showed that a significant portion of change in LD could be explained from the information contained in the genetic markers.

Technical Abstract: Conducting genomic selection in admixed populations is challenging and its accuracy in this case largely depends on the persistence of linkage disequilibrium between markers and QTLs. Inferring linkage disequilibrium between markers and QTLs could be important in understanding the change of SNP marker effects across different breeds. Predicting the change in linkage disequilibrium between markers and QTLs across two divergent breeds was explored using information from the genotype data. Two different models (M1, M2) that differ in the definition of the explanatory variables were used to infer the level of LD between SNP markers and QTLs using all markers in the panel or windows of fixed number of markers. Three simulation scenarios were conducted using different number of SNPs and QTLs. In the first scenario, the resulting coefficient of determination (R2) was 0.65 and 0.52 using M1 and M2, respectively. In the second scenario, average R2 equaled 0.12 using all markers in the panel and 0.25 using 100 marker windows. Across the three simulation scenarios, it was clear that a significant portion of the variation in the change in LD between SNP markers and QTLs could be explained by information already available in the observed SNP marker data.