A Weighted Genomic Relationship Matrix Based on Fixation Index (FST) Prioritized SNPs for Genomic Selection

A dramatic increase in the density of marker panels has been expected to increase the accuracy of genomic selection (GS), unfortunately, little to no improvement has been observed. By including all variants in the association model, the dimensionality of the problem should be dramatically increased, and it could undoubtedly reduce the statistical power. Using all Single nucleotide polymorphisms (SNPs) to compute the genomic relationship matrix (G) does not necessarily increase accuracy as the additive relationships can be accurately estimated using a much smaller number of markers. Due to these limitations, variant prioritization has become a necessity to improve accuracy. The fixation index (FST) as a measure of population differentiation has been used to identify genome segments and variants under selection pressure. Using prioritized variants has increased the accuracy of GS. Additionally, FST can be used to weight the relative contribution of prioritized SNPs in computing G. In this study, relative weights based on FST scores were developed and incorporated into the calculation of G and their impact on the estimation of variance components and accuracy was assessed. The results showed that prioritizing SNPs based on their FST scores resulted in an increase in the genetic similarity between training and validation animals and improved the accuracy of GS by more than 5%.

Download Full-text

Using the genomic relationship matrix to predict the accuracy of genomic selection

Journal of Animal Breeding and Genetics ◽

10.1111/j.1439-0388.2011.00964.x ◽

2011 ◽

Vol 128 (6) ◽

pp. 409-421 ◽

Cited By ~ 170

Author(s):

M.E. Goddard ◽

B.J. Hayes ◽

T.H.E. Meuwissen

Keyword(s):

Genomic Selection ◽

Genomic Relationship Matrix ◽

Relationship Matrix ◽

Genomic Relationship

Download Full-text

Accuracy of genomic BLUP when considering a genomic relationship matrix based on the number of the largest eigenvalues: a simulation study

Genetics Selection Evolution ◽

10.1186/s12711-019-0516-0 ◽

2019 ◽

Vol 51 (1) ◽

Cited By ~ 3

Author(s):

Ivan Pocrnic ◽

Daniela A. L. Lourenco ◽

Yutaka Masuda ◽

Ignacy Misztal

Keyword(s):

Genomic Selection ◽

Large Fraction ◽

Genomic Variation ◽

Eigenvalue Decomposition ◽

Genomic Relationship Matrix ◽

Relationship Matrix ◽

Genomic Relationship ◽

Genomic Information ◽

Phenotypic Information ◽

Largest Eigenvalues

Abstract Background The dimensionality of genomic information is limited by the number of independent chromosome segments (Me), which is a function of the effective population size. This dimensionality can be determined approximately by singular value decomposition of the gene content matrix, by eigenvalue decomposition of the genomic relationship matrix (GRM), or by the number of core animals in the algorithm for proven and young (APY) that maximizes the accuracy of genomic prediction. In the latter, core animals act as proxies to linear combinations of Me. Field studies indicate that a moderate accuracy of genomic selection is achieved with a small dataset, but that further improvement of the accuracy requires much more data. When only one quarter of the optimal number of core animals are used in the APY algorithm, the accuracy of genomic selection is only slightly below the optimal value. This suggests that genomic selection works on clusters of Me. Results The simulation included datasets with different population sizes and amounts of phenotypic information. Computations were done by genomic best linear unbiased prediction (GBLUP) with selected eigenvalues and corresponding eigenvectors of the GRM set to zero. About four eigenvalues in the GRM explained 10% of the genomic variation, and less than 2% of the total eigenvalues explained 50% of the genomic variation. With limited phenotypic information, the accuracy of GBLUP was close to the peak where most of the smallest eigenvalues were set to zero. With a large amount of phenotypic information, accuracy increased as smaller eigenvalues were added. Conclusions A small amount of phenotypic data is sufficient to estimate only the effects of the largest eigenvalues and the associated eigenvectors that contain a large fraction of the genomic information, and a very large amount of data is required to estimate the remaining eigenvalues that account for a limited amount of genomic information. Core animals in the APY algorithm act as proxies of almost the same number of eigenvalues. By using an eigenvalues-based approach, it was possible to explain why the moderate accuracy of genomic selection based on small datasets only increases slowly as more data are added.

Download Full-text

Simulation studies to optimize genomic selection in honey bees

Genetics Selection Evolution ◽

10.1186/s12711-021-00654-x ◽

2021 ◽

Vol 53 (1) ◽

Author(s):

Richard Bernstein ◽

Manuel Du ◽

Andreas Hoppe ◽

Kaspar Bienefeld

Keyword(s):

Genomic Selection ◽

Genetic Gain ◽

Honey Bees ◽

Reference Population ◽

Breeding Program ◽

Genomic Relationship Matrix ◽

Relationship Matrix ◽

Genomic Relationship ◽

Genomic Breeding ◽

Breeding Values

Abstract Background With the completion of a single nucleotide polymorphism (SNP) chip for honey bees, the technical basis of genomic selection is laid. However, for its application in practice, methods to estimate genomic breeding values need to be adapted to the specificities of the genetics and breeding infrastructure of this species. Drone-producing queens (DPQ) are used for mating control, and usually, they head non-phenotyped colonies that will be placed on mating stations. Breeding queens (BQ) head colonies that are intended to be phenotyped and used to produce new queens. Our aim was to evaluate different breeding program designs for the initiation of genomic selection in honey bees. Methods Stochastic simulations were conducted to evaluate the quality of the estimated breeding values. We developed a variation of the genomic relationship matrix to include genotypes of DPQ and tested different sizes of the reference population. The results were used to estimate genetic gain in the initial selection cycle of a genomic breeding program. This program was run over six years, and different numbers of genotyped queens per year were considered. Resources could be allocated to increase the reference population, or to perform genomic preselection of BQ and/or DPQ. Results Including the genotypes of 5000 phenotyped BQ increased the accuracy of predictions of breeding values by up to 173%, depending on the size of the reference population and the trait considered. To initiate a breeding program, genotyping a minimum number of 1000 queens per year is required. In this case, genetic gain was highest when genomic preselection of DPQ was coupled with the genotyping of 10–20% of the phenotyped BQ. For maximum genetic gain per used genotype, more than 2500 genotyped queens per year and preselection of all BQ and DPQ are required. Conclusions This study shows that the first priority in a breeding program is to genotype phenotyped BQ to obtain a sufficiently large reference population, which allows successful genomic preselection of queens. To maximize genetic gain, DPQ should be preselected, and their genotypes included in the genomic relationship matrix. We suggest, that the developed methods for genomic prediction are suitable for implementation in genomic honey bee breeding programs.

Download Full-text

Implications of simplified linkage equilibrium SNP simulation

10.1101/025619 ◽

2015 ◽

Author(s):

Sang Hong Lee

Keyword(s):

Covariance Structure ◽

Real Data ◽

Disease Status ◽

Genomic Relationship Matrix ◽

Relationship Matrix ◽

Linkage Equilibrium ◽

Genomic Relationship ◽

Single Nucleotide ◽

Realistic Simulation ◽

Data Analyses

In a recent paper published in PNAS (Golan et al. 2014), residual maximum likelihood (REML) seriously underestimated genetic variance explained by genomewide single nucleotide polymorphism when using a case-control design. It was concluded that Haseman–Elston regression (denoted as PCGC in their paper) should be used instead of REML. Their conclusions were based on results from simplified linkage equilibrium SNP simulation (SLES), which the authors acknowledged may be unrealistic. We found that their simulation, SLES, unrealistically inflated the correlation between the eigenvectors of the genomic relationship matrix and disease status to values that are rarely observed in real data analyses. With a more realistic simulation that the authors failed to carry out (as they noted in their paper), we showed that there was no such inflated correlation between the eigenvectors of the genomic relationship matrix and disease status. Because REML uses the eigensystem of covariance structure, the inflated correlation artefactually constrained its estimates. We compared SNP-heritabilities from SLES and a more realistic simulation, showing that there was a substantial difference between the REML estimates from the two simulation strategies. Finally, we presented that there was no difference between REML and PCGC in real data analyses. This pattern from real data results differed strikingly from the pattern in the simulation study of Golan et al. One needs to be cautious of results drawn from SLES.

Download Full-text

Genomic Relationship Matrix for Correcting Pedigree Errors in Breeding Populations: Impact on Genetic Parameters and Genomic Selection Accuracy

Crop Science ◽

10.2135/cropsci2012.12.0673 ◽

2014 ◽

Vol 54 (3) ◽

pp. 1115-1123 ◽

Cited By ~ 38

Author(s):

Patricio R. Munoz ◽

Marcio F. R. Resende ◽

Dudley A. Huber ◽

Tania Quesada ◽

Marcos D. V. Resende ◽

...

Keyword(s):

Genomic Selection ◽

Genetic Parameters ◽

Genomic Relationship Matrix ◽

Relationship Matrix ◽

Genomic Relationship ◽

Selection Accuracy ◽

Breeding Populations ◽

Pedigree Errors

Download Full-text

Improving the Efficiency of Genomic Selection in Chinese Simmental beef cattle

10.1101/022673 ◽

2015 ◽

Author(s):

Jiangwei Xia ◽

Yang Wu ◽

Huizhong Fang ◽

Wengang Zhang ◽

Yuxin Song ◽

...

Keyword(s):

Beef Cattle ◽

Genomic Relationship Matrix ◽

Relationship Matrix ◽

Nucleotide Polymorphisms ◽

Linear Unbiased Prediction ◽

Single Nucleotide ◽

Kinship Matrix ◽

Genome Wide ◽

Best Linear Unbiased ◽

Selection Of

Genomic selection is an accurate and efficient method of estimating genetic merits by using high-density genome-wide single nucleotide polymorphisms (SNPs).In this study, we investigate an approach to increase the efficiency of genomic prediction by using genome-wide markers. The approach is a feature selection based on genomic best linear unbiased prediction (GBLUP),which is a statistical method used to predict breeding values using SNPs for selection in animal and plant breeding. The objective of this study is the choice of kinship matrix for genomic best linear unbiased prediction (GBLUP).The G-matrix is using the information of genome-wide dense markers. We compare three kinds of kinships based on different combinations of centring and scaling of marker genotypes.And find a suitable kinship approach that adjusts for the resource population of Chinese Simmental beef cattle.Single nucleotide polymorphism (SNPs) can be used to estimate kinship matrix and individual inbreeding coefficients more accurately. So in our research a genomic relationship matrix was developed for 1059 Chinese Simmental beef cattle using 640000 single nucleotide polymorphisms and breeding values were estimated using phenotypes about Carcass weight and Sirloin weight. The number of SNPs needed to accurately estimate a genomic relationship matrix was evaluated in this population. Another aim of this study was to optimize the selection of markers and determine the required number of SNPs for estimation of kinship in the Chinese Simmental beef cattle. We find that the feature selection of GBLUP using Xu’s and the Astle and Balding’s kinships model performed similarly well, and were the best-performing methods in our study. Inbreeding and kinship matrix can be estimated with high accuracy using ≥12,000s in Chinese Simmental beef cattle.

Download Full-text

Assessing haplotype-based models for genomic evaluation in Holstein cattle

Canadian Journal of Animal Science ◽

10.1139/cjas-2018-0009 ◽

2018 ◽

Vol 98 (4) ◽

pp. 750-759 ◽

Cited By ~ 8

Author(s):

Z. Karimi ◽

M. Sargolzaei ◽

J.A.B. Robinson ◽

F.S. Schenkel

Keyword(s):

Breeding Value ◽

Validation Dataset ◽

Genomic Relationship Matrix ◽

Relationship Matrix ◽

Holstein Cattle ◽

Nucleotide Polymorphisms ◽

Genomic Relationship ◽

Wide Range ◽

High Heritability ◽

Prediction Reliability

A single-nucleotide polymorphisms-based genomic relationship matrix (GSNP) discriminate less identity by state from identity by descent (IBD) alleles compared with a multi-locus haplotype-based relationship matrix (GHAP), which can better capture IBD alleles and recent relationships. We aimed to compare the prediction reliability and prediction bias of genomic best linear unbiased prediction (GBLUP) using either GSNP or GHAP in Holstein cattle. Therefore, a total of 57 traits with a wide range of heritability values were analyzed. Classical validation tests were done using a validation dataset comprised of 50k genotype records of 561–669 proven bulls born in 2010–2011 with an official estimated breeding value (EBV) in 2016 and a training set of 5314–19 678 bulls born before 2010, depending on the trait. The method for building the genomic relationship matrix (G) had significant, but small effect on observed reliability (r2GEBV) (p < 0.0001) and bias (p < 0.0001). A significant interaction between G and the level of trait heritability on r2GEBV and bias was also observed (p < 0.0001). The small gains in r2GEBV and small reductions in the bias by using GHAPBLUP were increased when predicting moderate to high-heritability traits compared with low-heritability traits.

Download Full-text

Genomic selection using a realized genomic relationship matrix in a Pinus taeda L. cloned population

BMC Proceedings ◽

10.1186/1753-6561-5-s7-p60 ◽

2011 ◽

Vol 5 (Suppl 7) ◽

pp. P60 ◽

Cited By ~ 1

Author(s):

Jaime Zapata-Valenzuela ◽

Fikret Isik ◽

Christian Maltecca ◽

Jill Wegryzn ◽

David Neale ◽

...

Keyword(s):

Genomic Selection ◽

Pinus Taeda ◽

Genomic Relationship Matrix ◽

Relationship Matrix ◽

Genomic Relationship ◽

Pinus Taeda L

Download Full-text

PSVIII-27 A weighted genomic relationship matrix based on FST prioritized SNPs for genomic selection

Journal of Animal Science ◽

10.1093/jas/skz258.533 ◽

2019 ◽

Vol 97 (Supplement_3) ◽

pp. 262-262

Author(s):

Ling-Yun Chang ◽

Sajjad Toghiani ◽

E L Hamidi Hay ◽

Samuel E Aggrey ◽

Romdhane Rekaya

Keyword(s):

Genomic Selection ◽

Mixed Model ◽

Relative Weight ◽

Snp Markers ◽

Genomic Relationship Matrix ◽

Relationship Matrix ◽

Genomic Relationship ◽

Improve Accuracy ◽

Increase In Accuracy ◽

Generation Sequencing

Abstract Using low to moderate density SNP marker panels, a substantial increase in accuracy was achieved. The dramatic increase in the number of identified variants due to advances in next generation sequencing was expected to significantly increase the accuracy of genomic selection (GS). Unfortunately, little to no improvement was observed. For mixed model-based approaches, using all SNPs in the panel to compute the observed relationship matrix (G) will not increase accuracy as the additive relationships between individuals can be accurately estimated using a much smaller number of markers. Due to these limitations, variant prioritization has become a necessity to improve accuracy. Further, it has been shown that weighting SNPs when calculating G could be effective in improving the accuracy of GS. FST as a measure population differential has been successfully used to identify genome segments under selection pressure. Consequently, FST could be used to both prioritize SNPs and to derive their relative weight in the calculation of the genomic relationship matrix. A population of 15,000 animals genotyped for 400K SNP markers uniformly-distributed along 10 chromosomes was simulated. A trait with heritability 0.3 genetically controlled by two hundred QTL was generated. The top 20K SNPs based on their FST scores were used either alone or with the remaining 380K SNPs to compute G with or without weighting. When only the top 20K SNPs were used to compute G, two scenarios were considered: 1) equal weights for all SNPs or 2) weights proportional to the SNP FST scores. When all 400K SNP markers were used, different weighting scenarios were evaluated. The results clearly showed that prioritizing SNP markers based on their FST score and using the latter to compute relative weights has increased the genetic similarity between training and validations animals and resulted in more than 5% improvement in the accuracy of GS.

Download Full-text

Genomic Analyses of Globodera pallida, A Quarantine Agricultural Pathogen in Idaho

Pathogens ◽

10.3390/pathogens10030363 ◽

2021 ◽

Vol 10 (3) ◽

pp. 363

Author(s):

Sulochana K. Wasala ◽

Dana K. Howe ◽

Louise-Marie Dandurand ◽

Inga A. Zasada ◽

Dee R. Denver

Keyword(s):

Genetic Variation ◽

Potato Production ◽

Globodera Pallida ◽

Fixation Index ◽

Parasitic Nematodes ◽

Nucleotide Polymorphisms ◽

Single Nucleotide ◽

Genome Wide ◽

Field Samples ◽

Multiple Introduction

Globodera pallida is among the most significant plant-parasitic nematodes worldwide, causing major damage to potato production. Since it was discovered in Idaho in 2006, eradication efforts have aimed to contain and eradicate G. pallida through phytosanitary action and soil fumigation. In this study, we investigated genome-wide patterns of G. pallida genetic variation across Idaho fields to evaluate whether the infestation resulted from a single or multiple introduction(s) and to investigate potential evolutionary responses since the time of infestation. A total of 53 G. pallida samples (~1,042,000 individuals) were collected and analyzed, representing five different fields in Idaho, a greenhouse population, and a field in Scotland that was used for external comparison. According to genome-wide allele frequency and fixation index (Fst) analyses, most of the genetic variation was shared among the G. pallida populations in Idaho fields pre-fumigation, indicating that the infestation likely resulted from a single introduction. Temporal patterns of genome-wide polymorphisms involving (1) pre-fumigation field samples collected in 2007 and 2014 and (2) pre- and post-fumigation samples revealed nucleotide variants (SNPs, single-nucleotide polymorphisms) with significantly differentiated allele frequencies indicating genetic differentiation. This study provides insights into the genetic origins and adaptive potential of G. pallida invading new environments.

Download Full-text