Genomic prediction based on runs of homozygosity.

BACKGROUND: Genomic prediction is based on the accurate estimation of the genomic relationships among and between training animals and selection candidates in order to obtain accurate estimates of the genomic estimated breeding values (GEBV). Various methods have been used to predict GEBV based on population-wide linkage disequilibrium relationships (G IBS ) or sometimes on linkage analysis relationships (G LA ). Here, we propose a novel method to predict GEBV based on a genomic relationship matrix using runs of homozygosity (G ROH ). Runs of homozygosity were used to derive probabilities of multi-locus identity by descent chromosome segments. The accuracy and bias of the prediction of GEBV using G ROH were compared to those using G IBS and G LA . Comparisons were performed using simulated datasets derived from a random pedigree and a real pedigree of Italian Brown Swiss bulls. The comparison of accuracies of GEBV was also performed on data from 1086 Italian Brown Swiss dairy cattle. RESULTS: Simulations with various thresholds of minor allele frequency for markers and quantitative trait loci showed that G ROH achieved consistently more accurate GEBV (0 to 4% points higher) than G IBS and G LA . The bias of GEBV prediction for simulated data was higher based on the real pedigree than based on a random pedigree. In the analyses with real data, G ROH and G LA had similar accuracies. However, G LA achieved a higher accuracy when the prediction was done on the youngest animals. The G IBS matrices calculated with and without standardized marker genotypes resulted in similar accuracies. CONCLUSIONS: The present study proposes G ROH as a novel method to estimate genomic relationship matrices and predict GEBV based on runs of homozygosity and shows that it can result in higher or similar accuracies of GEBV prediction than G LA , except for the real data analysis with validation of young animals. Compared to G IBS , G ROH resulted in more accurate GEBV predictions.

Advantages of using molecular coancestry in the removal of introgressed genetic material.

BACKGROUND: When introgression of undesired exogenous genetic material occurs in a population intended to remain pure, actions are necessary to recover the original background. It has been shown that genome-wide information can replace pedigree information for different objectives and is a valuable tool in the fields of genetic conservation and breeding. In this simulation study, molecular information provided by 50 000 SNP was used to minimise the molecular coancestry between individuals of an admixed population and the foreign individuals that originally introgressed a native population in order to remove the exogenous DNA. RESULTS: This management method, which detects the 'purest' individuals to be used as parents for the next generation, allowed recovery of the native genetic background to a great extent in all simulated scenarios. However, it also caused an increase in inbreeding larger than expected because of the lower number of individuals selected as parents and the higher coancestry between them. In scenarios involving several introgression events the method was more efficient than in those involving a single introgression event because part of the genetic information was mixed with the native genetic material for a shorter period. CONCLUSIONS: Genome-wide information can be used to identify the purest individuals via the minimisation of molecular coancestry between individuals of the admixed and exogenous populations. Removal of the undesired genetic material is more efficient with a molecular-based approach than with a pedigree-based approach.

The importance of identity-by-state information for the accuracy of genomic selection.

BACKGROUND: It is commonly assumed that prediction of genome-wide breeding values in genomic selection is achieved by capitalizing on linkage disequilibrium between markers and QTL but also on genetic relationships. Here, we investigated the reliability of predicting genome-wide breeding values based on population-wide linkage disequilibrium information, based on identity-by-descent relationships within the known pedigree, and to what extent linkage disequilibrium information improves predictions based on identity-by-descent genomic relationship information. METHODS: The study was performed on milk, fat, and protein yield, using genotype data on 35 706 SNP and deregressed proofs of 1086 Italian Brown Swiss bulls. Genome-wide breeding values were predicted using a genomic identity-by-state relationship matrix and a genomic identity-by-descent relationship matrix (averaged over all marker loci). The identity-by-descent matrix was calculated by linkage analysis using one to five generations of pedigree data. RESULTS: We showed that genome-wide breeding values prediction based only on identity-by-descent genomic relationships within the known pedigree was as or more reliable than that based on identity-by-state, which implicitly also accounts for genomic relationships that occurred before the known pedigree. Furthermore, combining the two matrices did not improve the prediction compared to using identity-by-descent alone. Including different numbers of generations in the pedigree showed that most of the information in genome-wide breeding values prediction comes from animals with known common ancestors less than four generations back in the pedigree. CONCLUSIONS: Our results show that, in pedigreed breeding populations, the accuracy of genome-wide breeding values obtained by identity-by-descent relationships was not improved by identity-by-state information. Although, in principle, genomic selection based on identity-by-state does not require pedigree data, it does use the available pedigree structure. Our findings may explain why the prediction equations derived for one breed may not predict accurate genome-wide breeding values when applied to other breeds, since family structures differ among breeds.

The complete linkage disequilibrium test: a test that points to causative mutations underlying quantitative traits.

BACKGROUND: Genetically, SNP that are in complete linkage disequilibrium with the causative SNP cannot be distinguished from the causative SNP. The Complete Linkage Disequilibrium (CLD) test presented here tests whether a SNP is in complete LD with the causative mutation or not. The performance of the CLD test is evaluated in 1000 simulated datasets. METHODS: The CLD test consists of two steps i.e. analysis I and analysis II. Analysis I consists of an association analysis of the investigated region. The log-likelihood values from analysis I are next ranked in descending order and in analysis II the CLD test evaluates differences in log-likelihood ratios between the best and second best markers. Under the null-hypothesis distribution, the best SNP is in greater LD with the QTL than the second best, while under the alternative-CLD-hypothesis, the best SNP is alike-in-state with the QTL. To find a significance threshold, the test was also performed on data excluding the causative SNP. The 5th, 10th and 50th highest TCLD value from 1000 replicated analyses were used to control the type-I-error rate of the test at p = 0.005, p = 0.01 and p = 0.05, respectively. RESULTS: In a situation where the QTL explained 48% of the phenotypic variance analysis I detected a QTL in 994 replicates (p = 0.001), where 972 were positioned in the correct QTL position. When the causative SNP was excluded from the analysis, 714 replicates detected evidence of a QTL (p = 0.001). In analysis II, the CLD test confirmed 280 causative SNP from 1000 simulations (p = 0.05), i.e. power was 28%. When the effect of the QTL was reduced by doubling the error variance, the power of the test reduced relatively little to 23%. When sequence data were used, the power of the test reduced to 16%. All SNP that were confirmed by the CLD test were positioned in the correct QTL position. CONCLUSIONS: The CLD test can provide evidence for a causative SNP, but its power may be low in situations with closely linked markers. In such situations, also functional evidence will be needed to definitely conclude whether the SNP is causative or not.