Limits...
Genomic selection and association mapping in rice (Oryza sativa): effect of trait genetic architecture, training population composition, marker number and statistical model on accuracy of rice genomic selection in elite, tropical rice breeding lines.

Spindel J, Begum H, Akdemir D, Virk P, Collard B, Redoña E, Atlin G, Jannink JL, McCouch SR - PLoS Genet. (2015)

Bottom Line: The training population, statistical method used to build the GS model, number of markers, and trait were varied to determine their effect on prediction accuracy.For plant height, in which four mid-sized QTL were identified by GWAS, random forest produced the most consistently accurate GS models.Our results suggest that GS, informed by GWAS interpretations of genetic architecture and population structure, could become an effective tool for increasing the efficiency of rice breeding as the costs of genotyping continue to decline.

View Article: PubMed Central - PubMed

Affiliation: Department of Plant Breeding and Genetics, Cornell University, Ithaca, New York, United States of America.

ABSTRACT
Genomic Selection (GS) is a new breeding method in which genome-wide markers are used to predict the breeding value of individuals in a breeding population. GS has been shown to improve breeding efficiency in dairy cattle and several crop plant species, and here we evaluate for the first time its efficacy for breeding inbred lines of rice. We performed a genome-wide association study (GWAS) in conjunction with five-fold GS cross-validation on a population of 363 elite breeding lines from the International Rice Research Institute's (IRRI) irrigated rice breeding program and herein report the GS results. The population was genotyped with 73,147 markers using genotyping-by-sequencing. The training population, statistical method used to build the GS model, number of markers, and trait were varied to determine their effect on prediction accuracy. For all three traits, genomic prediction models outperformed prediction based on pedigree records alone. Prediction accuracies ranged from 0.31 and 0.34 for grain yield and plant height to 0.63 for flowering time. Analyses using subsets of the full marker set suggest that using one marker every 0.2 cM is sufficient for genomic selection in this collection of rice breeding materials. RR-BLUP was the best performing statistical method for grain yield where no large effect QTL were detected by GWAS, while for flowering time, where a single very large effect QTL was detected, the non-GS multiple linear regression method outperformed GS models. For plant height, in which four mid-sized QTL were identified by GWAS, random forest produced the most consistently accurate GS models. Our results suggest that GS, informed by GWAS interpretations of genetic architecture and population structure, could become an effective tool for increasing the efficiency of rice breeding as the costs of genotyping continue to decline.

Show MeSH

Related in: MedlinePlus

Example of irrigated rice breeding pipeline that incorporates genomic selection.Parents are selected and crossed to create an F1 population. ∼20,000 F1 lines are fixed over 7–8 generations with selection of families for heritable traits with ∼25% of the pedigree lines eventually selected for entry into the observational yield trial (OYT). GEBVs can be used at two or more generations during fixation as resources permit to perform selection. Here we propose using GEBVs at the F3 and F6 generations. GEBVs are also used to select the fixed lines from the F8 to advance to the OYT. The top lines advanced to the OYT based on GEBV are cycled back into the crossing block in order to continue to improve the population. From the OYT, the best performing lines based on phenotype are advanced to the replicated yield trials (RYT), and the best performing lines from the RYT are advanced to the multi-environment trials (MET). Lines from the MET are then selected based on GEBV as parents for the next generation of recurrent selection. Models are built at each stage in which GEBVs are used for selection based on a subset of the lines in the population (∼300 individuals representing different families) that are both genotyped and phenotyped to form the training set. The rest of the individuals in the population are genotyped only in order to calculate GEBVs.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4334555&req=5

pgen.1004982.g002: Example of irrigated rice breeding pipeline that incorporates genomic selection.Parents are selected and crossed to create an F1 population. ∼20,000 F1 lines are fixed over 7–8 generations with selection of families for heritable traits with ∼25% of the pedigree lines eventually selected for entry into the observational yield trial (OYT). GEBVs can be used at two or more generations during fixation as resources permit to perform selection. Here we propose using GEBVs at the F3 and F6 generations. GEBVs are also used to select the fixed lines from the F8 to advance to the OYT. The top lines advanced to the OYT based on GEBV are cycled back into the crossing block in order to continue to improve the population. From the OYT, the best performing lines based on phenotype are advanced to the replicated yield trials (RYT), and the best performing lines from the RYT are advanced to the multi-environment trials (MET). Lines from the MET are then selected based on GEBV as parents for the next generation of recurrent selection. Models are built at each stage in which GEBVs are used for selection based on a subset of the lines in the population (∼300 individuals representing different families) that are both genotyped and phenotyped to form the training set. The rest of the individuals in the population are genotyped only in order to calculate GEBVs.

Mentions: While genomic selection has yet to be integrated into applied breeding programs in rice as it has in maize and wheat, it would be feasible to undertake small pilot programs within specific rice breeding programs, especially for irrigated rice where growing environments are generally more uniform. Such pilot programs are needed, in particular, to determine when and how to incorporate genomic selection into existing breeding programs. An example of an irrigated rice breeding pipeline that incorporates genomic selection is presented in Fig. 2. Parents are selected and crossed and the resulting F1 progeny fixed over seven generations with selection of families for heritable traits. Traditionally, selection during pedigree line fixation would be based only on phenotype. Here, we propose incorporating selection based on GEBV at least once during fixation, as resources allow. Early generation GEBV-based selection would help to avoid eliminating families that carry beneficial rare or recessive alleles and would increase the proportion of top performers that are advanced to the observational yield trials (OYT). Late-generation selection based on GEBVs could be used to select fixed lines to advance to the OYT. The top lines advanced to the OYT based on GEBV could be used simultaneously as parents of the next generation of breeding (Fig. 2).


Genomic selection and association mapping in rice (Oryza sativa): effect of trait genetic architecture, training population composition, marker number and statistical model on accuracy of rice genomic selection in elite, tropical rice breeding lines.

Spindel J, Begum H, Akdemir D, Virk P, Collard B, Redoña E, Atlin G, Jannink JL, McCouch SR - PLoS Genet. (2015)

Example of irrigated rice breeding pipeline that incorporates genomic selection.Parents are selected and crossed to create an F1 population. ∼20,000 F1 lines are fixed over 7–8 generations with selection of families for heritable traits with ∼25% of the pedigree lines eventually selected for entry into the observational yield trial (OYT). GEBVs can be used at two or more generations during fixation as resources permit to perform selection. Here we propose using GEBVs at the F3 and F6 generations. GEBVs are also used to select the fixed lines from the F8 to advance to the OYT. The top lines advanced to the OYT based on GEBV are cycled back into the crossing block in order to continue to improve the population. From the OYT, the best performing lines based on phenotype are advanced to the replicated yield trials (RYT), and the best performing lines from the RYT are advanced to the multi-environment trials (MET). Lines from the MET are then selected based on GEBV as parents for the next generation of recurrent selection. Models are built at each stage in which GEBVs are used for selection based on a subset of the lines in the population (∼300 individuals representing different families) that are both genotyped and phenotyped to form the training set. The rest of the individuals in the population are genotyped only in order to calculate GEBVs.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4334555&req=5

pgen.1004982.g002: Example of irrigated rice breeding pipeline that incorporates genomic selection.Parents are selected and crossed to create an F1 population. ∼20,000 F1 lines are fixed over 7–8 generations with selection of families for heritable traits with ∼25% of the pedigree lines eventually selected for entry into the observational yield trial (OYT). GEBVs can be used at two or more generations during fixation as resources permit to perform selection. Here we propose using GEBVs at the F3 and F6 generations. GEBVs are also used to select the fixed lines from the F8 to advance to the OYT. The top lines advanced to the OYT based on GEBV are cycled back into the crossing block in order to continue to improve the population. From the OYT, the best performing lines based on phenotype are advanced to the replicated yield trials (RYT), and the best performing lines from the RYT are advanced to the multi-environment trials (MET). Lines from the MET are then selected based on GEBV as parents for the next generation of recurrent selection. Models are built at each stage in which GEBVs are used for selection based on a subset of the lines in the population (∼300 individuals representing different families) that are both genotyped and phenotyped to form the training set. The rest of the individuals in the population are genotyped only in order to calculate GEBVs.
Mentions: While genomic selection has yet to be integrated into applied breeding programs in rice as it has in maize and wheat, it would be feasible to undertake small pilot programs within specific rice breeding programs, especially for irrigated rice where growing environments are generally more uniform. Such pilot programs are needed, in particular, to determine when and how to incorporate genomic selection into existing breeding programs. An example of an irrigated rice breeding pipeline that incorporates genomic selection is presented in Fig. 2. Parents are selected and crossed and the resulting F1 progeny fixed over seven generations with selection of families for heritable traits. Traditionally, selection during pedigree line fixation would be based only on phenotype. Here, we propose incorporating selection based on GEBV at least once during fixation, as resources allow. Early generation GEBV-based selection would help to avoid eliminating families that carry beneficial rare or recessive alleles and would increase the proportion of top performers that are advanced to the observational yield trials (OYT). Late-generation selection based on GEBVs could be used to select fixed lines to advance to the OYT. The top lines advanced to the OYT based on GEBV could be used simultaneously as parents of the next generation of breeding (Fig. 2).

Bottom Line: The training population, statistical method used to build the GS model, number of markers, and trait were varied to determine their effect on prediction accuracy.For plant height, in which four mid-sized QTL were identified by GWAS, random forest produced the most consistently accurate GS models.Our results suggest that GS, informed by GWAS interpretations of genetic architecture and population structure, could become an effective tool for increasing the efficiency of rice breeding as the costs of genotyping continue to decline.

View Article: PubMed Central - PubMed

Affiliation: Department of Plant Breeding and Genetics, Cornell University, Ithaca, New York, United States of America.

ABSTRACT
Genomic Selection (GS) is a new breeding method in which genome-wide markers are used to predict the breeding value of individuals in a breeding population. GS has been shown to improve breeding efficiency in dairy cattle and several crop plant species, and here we evaluate for the first time its efficacy for breeding inbred lines of rice. We performed a genome-wide association study (GWAS) in conjunction with five-fold GS cross-validation on a population of 363 elite breeding lines from the International Rice Research Institute's (IRRI) irrigated rice breeding program and herein report the GS results. The population was genotyped with 73,147 markers using genotyping-by-sequencing. The training population, statistical method used to build the GS model, number of markers, and trait were varied to determine their effect on prediction accuracy. For all three traits, genomic prediction models outperformed prediction based on pedigree records alone. Prediction accuracies ranged from 0.31 and 0.34 for grain yield and plant height to 0.63 for flowering time. Analyses using subsets of the full marker set suggest that using one marker every 0.2 cM is sufficient for genomic selection in this collection of rice breeding materials. RR-BLUP was the best performing statistical method for grain yield where no large effect QTL were detected by GWAS, while for flowering time, where a single very large effect QTL was detected, the non-GS multiple linear regression method outperformed GS models. For plant height, in which four mid-sized QTL were identified by GWAS, random forest produced the most consistently accurate GS models. Our results suggest that GS, informed by GWAS interpretations of genetic architecture and population structure, could become an effective tool for increasing the efficiency of rice breeding as the costs of genotyping continue to decline.

Show MeSH
Related in: MedlinePlus