Source: UNIVERSITY OF GEORGIA submitted to NRP
STRATEGIES TO IMPROVE THE ACCURACY OF ESTIMATING GENETIC MERIT IN THE ABSENCE OF GENOMIC DATA BY MODELING THE MENDELIAN SAMPLING
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
HATCH
Reporting Frequency
Annual
Accession No.
1009463
Grant No.
(N/A)
Cumulative Award Amt.
(N/A)
Proposal No.
(N/A)
Multistate No.
(N/A)
Project Start Date
Apr 23, 2016
Project End Date
Mar 31, 2021
Grant Year
(N/A)
Program Code
[(N/A)]- (N/A)
Recipient Organization
UNIVERSITY OF GEORGIA
200 D.W. BROOKS DR
ATHENS,GA 30602-5016
Performing Department
Animal & Dairy Science
Non Technical Summary
Large scale genotyping for single nucleotide polymorphisms has provided an unprecedented resource to study associations between traits and genomic variation. Undeniable success has been observed in identifying relevant associations for human diseases, and livestock traits. In livestock applications, the benefit of using genomic information is not limited to the genetic dissection of complex traits, but also to enhance the estimation of breeding values (BV) through genomic selection (GS). In dairy cattle, accuracy of the genomically estimated BVs is 30% to 70% higher than their counterparts obtained using BLUP. Superiority of GS compared to BLUP is due to the use of the realized relationships rather than the expected relationships. Unfortunately, in any livestock species only a small fraction of the population is currently genotyped (less than 1%), thus limiting the full potential of GS. Given the unlikely scenario that a substantial portion of the population will be genotyped anytime soon, it is of great theoretical and practical importance to infer or at least approximate the genomic relationship matrix, in the absence of genomic data. We propose a new method to enhance the additive relationship matrix in absence of genomic data. We will evaluate its effectiveness, and develop the needed software for its implementation. To reach that goal, we will leverage information in the already available phenotypic records and the empirical variation in co-ancestry coefficients to infer the actual additive relationships.
Animal Health Component
20%
Research Effort Categories
Basic
80%
Applied
20%
Developmental
(N/A)
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
3033499108120%
3033399108140%
3033299108140%
Knowledge Area
303 - Genetic Improvement of Animals;

Subject Of Investigation
3499 - Dairy cattle, general/other; 3299 - Poultry, general/other; 3399 - Beef cattle, general/other;

Field Of Science
1081 - Breeding;
Goals / Objectives
Our long term goal is to develop methodologies to improve the accuracy of estimating genetic merit with or without genomic data. We propose to assess the possibility of enhancing the expected average relationship matrix between animals using phenotypic data and co-ancestry coefficient variation as prior information in absence of genomic data. We will develop the necessary analytical and computational tools to make full use of already available information, and evaluate its potential improvement of genetic evaluation. To achieve these goals the following specific objectives will be implemented:Develop a joint prior using phenotypic data and co-ancestry coefficient variation and assess its ability to model the Mendelian sampling and to enhance the expected relationship matrix in absence of genomic dataCompare the performance of the newly derive matrix with the expected and realized relationships matrices in predicting the genetic merit
Project Methods
Due to inability to type too many equations, please see the attached complete pdf file is the "proposal" section

Progress 04/23/16 to 03/31/21

Outputs
Target Audience:The project is basic in nature with the overall aim to improve the accuracy of genetic evaluation of livestock and poultry. The primary target audience are scientists at academic institutions, and researchers at animal genetic improvement companies. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Three graduate students were trained during the implementation of the project. Two PhD students have already graduated and the third one is finishing her dissertation work. How have the results been disseminated to communities of interest?The results were primarily disseminated via referred journal articles, abstracts, and presentations at professional meetings What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? The overall aim of the project was toimprove the accuracy of genetic evaluation with or without genomic data. Our results showed that predicting the genomic relationships using correlated phenotypes has low accuracy to be used in practical applications. However, the prioritization of a small subset of SNP markers using population differentiation parameters such as Fst could increase the accuracy of genomic selections. In fact, we showed using simulated and real data that marker prioritization yielded more accurate results compared to regression based and single step approaches.It safe to conclude that: The Mendelian sampling cannot be accurately approximated using correlated phenotypes even when the heritabilities are moderate to high A small subset of correctly prioritize SNP markers (in true linkage disequilibrium with QTL) could significantly improve the accuracy of genomic prediction The set of prioritized SNP markers could be used directly in the association model or to construct the genomic relationship matrix for GBLUP type implementations In the presence of very complex traits, marker prioritization methods are less efficient

Publications

  • Type: Journal Articles Status: Under Review Year Published: 2021 Citation: Ling, A., El Hamidi Hay, Samuel E. Aggrey, and Romdhane Rekaya. Dissection of the impact of prioritized QTL-linked and -unlinked SNP markers on the accuracy of genomic selection. BMC Genetics (Under Review)
  • Type: Conference Papers and Presentations Status: Published Year Published: 2020 Citation: Aggrey, S. E., and R. Rekaya (2020). Genomic prediction using prioritized SNP markers: Application in Poultry. Poultry Breeding Symposium, Wageningen University, Netherlands; March 06, 2020
  • Type: Conference Papers and Presentations Status: Published Year Published: 2020 Citation: Aggrey, S. E., S. Toghiani, L. Y. Chang, and R. Rekaya. (2020). Improving accuracy of genomic prediction using a selected small set of prioritized SNP markers
  • Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: Sumreddee, P., S. Toghiani, S. E. Aggrey, and R. Rekaya. Joint genome wide association analysis of continuous and discrete traits. ASAS-CSAS Annual Meeting. Baltimore, MD from July 8 to July 12, 2017.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: Chang, L.Y., S. Toghiani, S. E. Aggrey, and R. Rekaya. 2017. Increasing accuracy of genomic selection in presence of high density marker panels through the prioritization of relevant polymorphisms. ASAS-CSAS Annual Meeting. Baltimore, MD from July 8 to July 12, 2017.
  • Type: Journal Articles Status: Published Year Published: 2017 Citation: Toghiani, S., L. Y. Chang, A. Ling, S. E. Aggrey, R. Rekaya. (2017). Genomic differentiation as a tool for Single Nucleotide Polymorphism prioritization for Genome wide association and phenotype prediction in livestock. Livestock Science 205: 24-30.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: Toghiani, S., L. Y. Chang, S. E. Aggrey, R. Rekaya. 2017. A hybrid of prioritized SNP and polygenetic effect method for implementation of genomic selection. J. Anim. Sci. 95 (supplement 4):92 (Abstr).
  • Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: Chang, L. Y., S. Toghiani, S. E. Aggrey, R. Rekaya. 2017. Increasing accuracy of genomic selection in presence of high density marker panels through the prioritization of relevant polymorphisms. Anim. Sci. 95 (supplement 4):91 (Abstr).
  • Type: Journal Articles Status: Published Year Published: 2018 Citation: Chang LY, Toghiani S, Ling A, Aggrey SE, Rekaya R. 2018. High density marker panels, SNPs prioritizing and accuracy of genomic selection. BMC Genetics 2018; 19:4.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Ling, A., S. E. Aggrey, and R. rekaya. Mendelian sampling, QTL similarity, and accuracy of genomic selection. JOURNAL OF ANIMAL SCIENCE 97:51-52 01 Dec 2019
  • Type: Conference Papers and Presentations Status: Published Year Published: 2018 Citation: R Rekaya, S Toghiani, P Sumreddee, A Ling, S Aggrey. 2018. Multivariate genome wide association for continuous and discrete responses using multivariate Bernoulli prior. J. Anim. Sci. 96 (Suppl 3):126-126. DOI:10.1093/jas/sky404.277
  • Type: Journal Articles Status: Published Year Published: 2019 Citation: Chang, L. -Y., Toghiani, S., Aggrey, S. E., & Rekaya, R. (2019). Increasing accuracy of genomic selection in presence of high density marker panels through the prioritization of relevant polymorphisms. BMC Genet, 20(1), 21. doi:10.1186/s12863-019-0720-5
  • Type: Journal Articles Status: Published Year Published: 2019 Citation: Chang, L. -Y., Toghiani, S., Hay, E. H., Aggrey, S. E., & Rekaya, R. (2019). A Weighted Genomic Relationship Matrix Based on Fixation Index (FST) Prioritized SNPs for Genomic Selection. Genes, 10(11), 922. doi:10.3390/genes10110922
  • Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Toghiani, S., L. Y. Chang, E. H. Hay, A. J. Roberts, S. E. Aggrey, and R. Rekaya. (2019). A hybrid model for genomic selection using prioritized SNPs based on FST scores in the presence of non-genotyped animals. J. Anim. Sci. 97 (Suppl )


Progress 10/01/19 to 09/30/20

Outputs
Target Audience:The target audience of the project are scientists and researchers at academic institutions and livestock companies Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Two graduate students were trained on the projects. How have the results been disseminated to communities of interest?Journal articles, abstracts and professional meeting presentation were the primary dissemination methods of the results What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? We assessed the impact of QTL-linked and unlinked SNP markers on the accuray of genomic selection.Identification and exclusion of markers that have little to no relevance to the trait of interest may significantly increase genomic prediction accuracies. The population statistic FSTwasan efficient and effective tool for preselection of trait-relevant markers.

Publications

  • Type: Journal Articles Status: Accepted Year Published: 2021 Citation: 157. #Ling, A., El Hamidi Hay, Samuel E. Aggrey, and Romdhane Rekaya. Dissection of the impact of prioritized QTL-linked and -unlinked SNP markers on the accuracy of genomic selection. BMC Genetics (Under Review)
  • Type: Conference Papers and Presentations Status: Published Year Published: 2020 Citation: Aggrey, S. E., S. Toghiani, L. Y. Chang, and R. Rekaya (2020). Improving accuracy of genomic prediction using a selected small set of prioritized SNP markers
  • Type: Conference Papers and Presentations Status: Published Year Published: 2020 Citation: Aggrey, S. E., and R. Rekaya. Genomic prediction using prioritized SNP markers: Application in Poultry. Poultry Breeding Symposium, Wageningen University, Netherlands; March 06, 2020


Progress 10/01/18 to 09/30/19

Outputs
Target Audience:Activities carried out under this project during this reporting period targeted mainly Animal breeders, scientists in the field of quantitative genetics and graduate students Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Two PhD students are being training based on activities related to this project. The training included development of algorithms and computer software needed for the implementation of some of the analyses. Additionally, the students were trained in several aspects of genomic selection and its application in the field of animal breeding and genetics. How have the results been disseminated to communities of interest?2. Presentation at industry and professional meeting and workshops (1 meeting abstract) What do you plan to do during the next reporting period to accomplish the goals?Our plan for the next reporting period is: 1. To further reduce the computational costs for the implementation of the hybrid model in the presence of non-genotyped animals 2. To extend the hybrid model tomultiple traitsituations

Impacts
What was accomplished under these goals? Prioritization of SNP markers based on population differentiation scores (FST) has led to an increase in the accuracy of genomic selection. Unfortunately, this information is not available for non-genotyped animals. For the latter, a prior with asymmetric covariance matrix was used to overcome the limitations imposed by the lack of genotype information. The proposed method was able to accommodate non-genotyped animals without the need to impute their missing SNP marker genotypes and resulted in substantial savings in computational costs. The method was implemented in a hybrid genomic evaluation model.

Publications

  • Type: Journal Articles Status: Published Year Published: 2019 Citation: Chang, L. -Y., Toghiani, S., Aggrey, S. E., & Rekaya, R. (2019). Increasing accuracy of genomic selection in presence of high density marker panels through the prioritization of relevant polymorphisms. BMC Genet, 20(1), 21. doi:10.1186/s12863-019-0720-5
  • Type: Journal Articles Status: Published Year Published: 2019 Citation: Chang, L. -Y., Toghiani, S., Hay, E. H., Aggrey, S. E., & Rekaya, R. (2019). A Weighted Genomic Relationship Matrix Based on Fixation Index (FST) Prioritized SNPs for Genomic Selection. Genes, 10(11), 922. doi:10.3390/genes10110922
  • Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: 127. Toghiani, S., L. Y. Chang, E. H. Hay, A. J. Roberts, S. E. Aggrey, and R. Rekaya. A hybrid model for genomic selection using prioritized SNPs based on FST scores in the presence of non-genotyped animals. J. Anim. Sci. 97 (Suppl )


Progress 10/01/17 to 09/30/18

Outputs
Target Audience:The target audience was mainly animal scientist and academicians, graduate students and the livestock and poultry industries Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Two graduate students were trained: 1. Sajjad Toghiani 2. Ling-Yun Chang How have the results been disseminated to communities of interest?Two journal articles and one conference abstract were published: Chang LY, Toghiani S, Ling A, Aggrey SE, Rekaya R. High density marker panels, SNPs prioritizing and accuracy of genomic selection. BMC Genetics 2018; 19:4. Chang LY, Toghiani S, Aggrey SE, Rekaya R. Increasing accuracy of genomic selection in the presence of high density marker panels through the prioritization of relevant polymorphisms. BMC Genetics (under review) R Rekaya, S Toghiani, P Sumreddee, A Ling, S Aggrey. 2018. Multivariategenome wideassociation for continuous and discrete responses using multivariate Bernoulli prior.J. Anim. Sci. 96 (Suppl 3):126-126. DOI:10.1093/jas/sky404.277 What do you plan to do during the next reporting period to accomplish the goals?1. Extend the approach to multiple trait scenarios including discrete responses

Impacts
What was accomplished under these goals? A hybrid method was developed to integrate prioritized SNP and polygenic components in the presence of non-genotyped animals. An asymmetric prior was used to accommodate the non-typed animals. Additionally, the prioritized SNPs were used to compute the genomic relationship matrix. Furthermore, the Fst score were used to assign relative weights to the different SNP markers in the panel

Publications

  • Type: Journal Articles Status: Published Year Published: 2018 Citation: Chang LY, Toghiani S, Ling A, Aggrey SE, Rekaya R. 2018. High density marker panels, SNPs prioritizing and accuracy of genomic selection. BMC Genetics 2018; 19:4.
  • Type: Journal Articles Status: Under Review Year Published: 2018 Citation: Chang LY, Toghiani S, Aggrey SE, Rekaya R. 2018. Increasing accuracy of genomic selection in presence of high density marker panels through the prioritization of relevant polymorphisms. BMC Genetics (under review)
  • Type: Conference Papers and Presentations Status: Published Year Published: 2018 Citation: 1. R Rekaya, S Toghiani, P Sumreddee, A Ling, S Aggrey. 2018. Multivariate genome wide association for continuous and discrete responses using multivariate Bernoulli prior. J. Anim. Sci. 96 (Suppl 3):126-126. DOI:10.1093/jas/sky404.277


Progress 10/01/16 to 09/30/17

Outputs
Target Audience:The target audience was mainly animal scientist and academicians, graduate students and the livestock and poultry industries Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Two graduate students were trained: 1. Sajjad Toghiani 2. Ling-Yun Chang How have the results been disseminated to communities of interest?One journal article and two conference abstracts were published: Toghiani, S., L. Y. Chang, A. Ling, S. E. Aggrey, R. Rekaya. (2017). Genomic differentiation as a tool for Single Nucleotide Polymorphism prioritization for Genome wide association and phenotype prediction in livestock. Livestock Science 205: 24-30. Toghiani, S., L. Y. Chang, S. E. Aggrey, R. Rekaya. 2017. A hybrid of prioritized SNP and polygenetic effect method for implementation of genomic selection. J. Anim. Sci. 95 (supplement 4):92 (Abstr). Chang, L. Y., S. Toghiani, S. E. Aggrey, R. Rekaya. 2017. Increasing accuracy of genomic selection in presence of high density marker panels through the prioritization of relevant polymorphisms. Anim. Sci. 95 (supplement 4):91 (Abstr). What do you plan to do during the next reporting period to accomplish the goals?Plan for the next reporting period: 1. Expand the developed method to large datasets in the presence of high density SNP marker information 2. Evaluate its effects on the enhancement of the genomic relationship matrix

Impacts
What was accomplished under these goals? We developed a new approach to prioritize SNPs based on Fst scores. Because genotypes of the prioritized SNPs will not be known for non-typed animals, we developed an asymmetric prior to impute the missing genotypes of non-typed animals using a Bayesian approach. The developed approach allowed the inclusion of non-genotyped animals in genomic selection at reasonable computational costs

Publications

  • Type: Journal Articles Status: Published Year Published: 2017 Citation: Toghiani, S., L. Y. Chang, A. Ling, S. E. Aggrey, R. Rekaya. (2017). Genomic differentiation as a tool for Single Nucleotide Polymorphism prioritization for Genome wide association and phenotype prediction in livestock. Livestock Science 205: 24-30.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: Toghiani, S., L. Y. Chang, S. E. Aggrey, R. Rekaya. 2017. A hybrid of prioritized SNP and polygenetic effect method for implementation of genomic selection. J. Anim. Sci. 95 (supplement 4):92 (Abstr).
  • Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: Chang, L. Y., S. Toghiani, S. E. Aggrey, R. Rekaya. 2017. Increasing accuracy of genomic selection in presence of high density marker panels through the prioritization of relevant polymorphisms. Anim. Sci. 95 (supplement 4):91 (Abstr).


Progress 04/23/16 to 09/30/16

Outputs
Target Audience:Scientists in the field of animal breeding and genetics Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Two graduate students are partially working on this project. How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals?1. Find the best approach and amount of information needed to infer the realized relationships 2. Evaluate the accuracy of the method under realistic production scenarios (number of traits, quality of the pedigree, etc.)

Impacts
What was accomplished under these goals? During the 5 month period, data simulations were carried out to investigae the optimium way of combining phenotypic information with co-ancestry coefficients. Relevanant factors such as number of traits, heritabilities, and genetic correlations are being investigated.

Publications

  • Type: Conference Papers and Presentations Status: Submitted Year Published: 2017 Citation: 1. Sumreddee, P., S. Toghiani, S. E. Aggrey, and R. Rekaya. Joint genome wide association analysis of continuous and discrete traits. ASAS-CSAS Annual Meeting. Baltimore, MD from July 8 to July 12, 2017. 2. Chang, L.Y., S. Toghiani, S. E. Aggrey, and R. Rekaya. 2017. Increasing accuracy of genomic selection in presence of high density marker panels through the prioritization of relevant polymorphisms. ASAS-CSAS Annual Meeting. Baltimore, MD from July 8 to July 12, 2017.