FiVeR - Repository of the Institute of Field and Vegetable Crops
Institute of Field and Vegetable Crops
    • English
    • Српски
    • Српски (Serbia)
  • English 
    • English
    • Serbian (Cyrillic)
    • Serbian (Latin)
  • Login
View Item 
  •   FiVeR
  • FiVeR
  • Radovi istraživača / Researchers' papers
  • View Item
  •   FiVeR
  • FiVeR
  • Radovi istraživača / Researchers' papers
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.

Exploring the performance of genomic prediction models for soybean yield using different validation approaches

Authorized Users Only
2019
Authors
Đorđević, Vuk
Ćeran, Marina
Miladinović, Jegor
Balešević-Tubić, Svetlana
Petrović, Kristina
Miladinov, Zlatica
Marinković, Jelena
Article (Published version)
Metadata
Show full item record
Abstract
Genomic selection is a valuable breeding tool that has a great potential for implementation in a real breeding program, as long as prediction model performance is carefully evaluated for each specific scenario. The performance of genomic prediction models has been commonly evaluated by standard cross-validation that can lead to an overestimation of the model performance, by using the same genetic material and their performances that were included in the model development. Besides cross-validation, this study explored the efficiency of yield prediction models for soybean (Glycine max (L.) Merr.) by using historical data for external model validation. Historical data represents a valuable source for evaluation of model performance, simulating the real breeding process. In general, results indicate a modest influence of statistical model and marker number on the prediction ability cross-validation and external validation. In both considerations, non-parametric random forest (RF) model sho...wed an overestimation of genomic estimated breeding values (GEBVs). Overall, genomic prediction ability for soybean yield for historical data across years was relatively high (0.60), implicating that the model has the potential to predict broad adaptation of breeding lines. The model, however, had variable ability to predict phenotypic performance in separate years, with especially high prediction ability in years not impacted by yield-limiting factors, when the genetic potential was fully achieved. General improvement of model performance in both cross-validation and external validation was achieved by increasing the phenotyping intensity that must reflect the target environment variability in terms of different climatological conditions.

Keywords:
Soybean / Genomic selection / Prediction models / Historical data / External validation
Source:
Molecular Breeding, 2019, 39, 5
Publisher:
  • Springer, Dordrecht
Funding / projects:
  • Interdisciplinary Approach to Development of New Soybean Varieties and Improvement of the Cultivation Practices and Seed Production (RS-31022)
  • Danube region (Deutsche Gesellschaft fur Internationale Zusammenarbeit GmbH)
  • Provincial Secretariat for Higher Education and Scientific Research, Autonomous Province of Vojvodina, Republic of Serbia, 114-451-2739/2016-01

DOI: 10.1007/s11032-019-0983-6

ISSN: 1380-3743

WoS: 000467653900002

Scopus: 2-s2.0-85065741333
[ Google Scholar ]
7
5
URI
http://fiver.ifvcns.rs/handle/123456789/1916
Collections
  • Radovi istraživača / Researchers' papers
Institution/Community
FiVeR
TY  - JOUR
AU  - Đorđević, Vuk
AU  - Ćeran, Marina
AU  - Miladinović, Jegor
AU  - Balešević-Tubić, Svetlana
AU  - Petrović, Kristina
AU  - Miladinov, Zlatica
AU  - Marinković, Jelena
PY  - 2019
UR  - http://fiver.ifvcns.rs/handle/123456789/1916
AB  - Genomic selection is a valuable breeding tool that has a great potential for implementation in a real breeding program, as long as prediction model performance is carefully evaluated for each specific scenario. The performance of genomic prediction models has been commonly evaluated by standard cross-validation that can lead to an overestimation of the model performance, by using the same genetic material and their performances that were included in the model development. Besides cross-validation, this study explored the efficiency of yield prediction models for soybean (Glycine max (L.) Merr.) by using historical data for external model validation. Historical data represents a valuable source for evaluation of model performance, simulating the real breeding process. In general, results indicate a modest influence of statistical model and marker number on the prediction ability cross-validation and external validation. In both considerations, non-parametric random forest (RF) model showed an overestimation of genomic estimated breeding values (GEBVs). Overall, genomic prediction ability for soybean yield for historical data across years was relatively high (0.60), implicating that the model has the potential to predict broad adaptation of breeding lines. The model, however, had variable ability to predict phenotypic performance in separate years, with especially high prediction ability in years not impacted by yield-limiting factors, when the genetic potential was fully achieved. General improvement of model performance in both cross-validation and external validation was achieved by increasing the phenotyping intensity that must reflect the target environment variability in terms of different climatological conditions.
PB  - Springer, Dordrecht
T2  - Molecular Breeding
T1  - Exploring the performance of genomic prediction models for soybean yield using different validation approaches
IS  - 5
VL  - 39
DO  - 10.1007/s11032-019-0983-6
UR  - conv_2709
ER  - 
@article{
author = "Đorđević, Vuk and Ćeran, Marina and Miladinović, Jegor and Balešević-Tubić, Svetlana and Petrović, Kristina and Miladinov, Zlatica and Marinković, Jelena",
year = "2019",
abstract = "Genomic selection is a valuable breeding tool that has a great potential for implementation in a real breeding program, as long as prediction model performance is carefully evaluated for each specific scenario. The performance of genomic prediction models has been commonly evaluated by standard cross-validation that can lead to an overestimation of the model performance, by using the same genetic material and their performances that were included in the model development. Besides cross-validation, this study explored the efficiency of yield prediction models for soybean (Glycine max (L.) Merr.) by using historical data for external model validation. Historical data represents a valuable source for evaluation of model performance, simulating the real breeding process. In general, results indicate a modest influence of statistical model and marker number on the prediction ability cross-validation and external validation. In both considerations, non-parametric random forest (RF) model showed an overestimation of genomic estimated breeding values (GEBVs). Overall, genomic prediction ability for soybean yield for historical data across years was relatively high (0.60), implicating that the model has the potential to predict broad adaptation of breeding lines. The model, however, had variable ability to predict phenotypic performance in separate years, with especially high prediction ability in years not impacted by yield-limiting factors, when the genetic potential was fully achieved. General improvement of model performance in both cross-validation and external validation was achieved by increasing the phenotyping intensity that must reflect the target environment variability in terms of different climatological conditions.",
publisher = "Springer, Dordrecht",
journal = "Molecular Breeding",
title = "Exploring the performance of genomic prediction models for soybean yield using different validation approaches",
number = "5",
volume = "39",
doi = "10.1007/s11032-019-0983-6",
url = "conv_2709"
}
Đorđević, V., Ćeran, M., Miladinović, J., Balešević-Tubić, S., Petrović, K., Miladinov, Z.,& Marinković, J.. (2019). Exploring the performance of genomic prediction models for soybean yield using different validation approaches. in Molecular Breeding
Springer, Dordrecht., 39(5).
https://doi.org/10.1007/s11032-019-0983-6
conv_2709
Đorđević V, Ćeran M, Miladinović J, Balešević-Tubić S, Petrović K, Miladinov Z, Marinković J. Exploring the performance of genomic prediction models for soybean yield using different validation approaches. in Molecular Breeding. 2019;39(5).
doi:10.1007/s11032-019-0983-6
conv_2709 .
Đorđević, Vuk, Ćeran, Marina, Miladinović, Jegor, Balešević-Tubić, Svetlana, Petrović, Kristina, Miladinov, Zlatica, Marinković, Jelena, "Exploring the performance of genomic prediction models for soybean yield using different validation approaches" in Molecular Breeding, 39, no. 5 (2019),
https://doi.org/10.1007/s11032-019-0983-6 .,
conv_2709 .

DSpace software copyright © 2002-2015  DuraSpace
About FiVeR | Send Feedback

OpenAIRERCUB
 

 

All of DSpaceInstitutions/communitiesAuthorsTitlesSubjectsThis institutionAuthorsTitlesSubjects

Statistics

View Usage Statistics

DSpace software copyright © 2002-2015  DuraSpace
About FiVeR | Send Feedback

OpenAIRERCUB