Exploring Deep Learning for Complex Trait Genomic Prediction in Polyploid Outcrossing Species
Genomic prediction (GP) is the procedure whereby the genetic merits of untested candidates are predicted using genome wide marker information. Although numerous examples of GP exist in plants and animals, applications to polyploid organisms are still scarce, partly due to limited genome resources an...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2020-02-01
|
| Series: | Frontiers in Plant Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/article/10.3389/fpls.2020.00025/full |
| _version_ | 1818924210196578304 |
|---|---|
| author | Laura M. Zingaretti Salvador Alejandro Gezan Luis Felipe V. Ferrão Luis F. Osorio Amparo Monfort Amparo Monfort Patricio R. Muñoz Vance M. Whitaker Miguel Pérez-Enciso Miguel Pérez-Enciso |
| author_facet | Laura M. Zingaretti Salvador Alejandro Gezan Luis Felipe V. Ferrão Luis F. Osorio Amparo Monfort Amparo Monfort Patricio R. Muñoz Vance M. Whitaker Miguel Pérez-Enciso Miguel Pérez-Enciso |
| author_sort | Laura M. Zingaretti |
| collection | DOAJ |
| description | Genomic prediction (GP) is the procedure whereby the genetic merits of untested candidates are predicted using genome wide marker information. Although numerous examples of GP exist in plants and animals, applications to polyploid organisms are still scarce, partly due to limited genome resources and the complexity of this system. Deep learning (DL) techniques comprise a heterogeneous collection of machine learning algorithms that have excelled at many prediction tasks. A potential advantage of DL for GP over standard linear model methods is that DL can potentially take into account all genetic interactions, including dominance and epistasis, which are expected to be of special relevance in most polyploids. In this study, we evaluated the predictive accuracy of linear and DL techniques in two important small fruits or berries: strawberry and blueberry. The two datasets contained a total of 1,358 allopolyploid strawberry (2n=8x=112) and 1,802 autopolyploid blueberry (2n=4x=48) individuals, genotyped for 9,908 and 73,045 single nucleotide polymorphism (SNP) markers, respectively, and phenotyped for five agronomic traits each. DL depends on numerous parameters that influence performance and optimizing hyperparameter values can be a critical step. Here we show that interactions between hyperparameter combinations should be expected and that the number of convolutional filters and regularization in the first layers can have an important effect on model performance. In terms of genomic prediction, we did not find an advantage of DL over linear model methods, except when the epistasis component was important. Linear Bayesian models were better than convolutional neural networks for the full additive architecture, whereas the opposite was observed under strong epistasis. However, by using a parameterization capable of taking into account these non-linear effects, Bayesian linear models can match or exceed the predictive accuracy of DL. A semiautomatic implementation of the DL pipeline is available at https://github.com/lauzingaretti/deepGP/. |
| first_indexed | 2024-12-20T02:21:42Z |
| format | Article |
| id | doaj.art-fe73c8137ccd476a8a6c0ab2bf8d1805 |
| institution | Directory Open Access Journal |
| issn | 1664-462X |
| language | English |
| last_indexed | 2024-12-20T02:21:42Z |
| publishDate | 2020-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Plant Science |
| spelling | doaj.art-fe73c8137ccd476a8a6c0ab2bf8d18052022-12-21T19:56:48ZengFrontiers Media S.A.Frontiers in Plant Science1664-462X2020-02-011110.3389/fpls.2020.00025506702Exploring Deep Learning for Complex Trait Genomic Prediction in Polyploid Outcrossing SpeciesLaura M. Zingaretti0Salvador Alejandro Gezan1Luis Felipe V. Ferrão2Luis F. Osorio3Amparo Monfort4Amparo Monfort5Patricio R. Muñoz6Vance M. Whitaker7Miguel Pérez-Enciso8Miguel Pérez-Enciso9Centre for Research in Agricultural Genomics (CRAG) CSIC-IRTA-UAB-UB, Campus UAB, Barcelona, SpainSchool of Forest Resources and Conservation, University of Florida, Gainesville, FL, United StatesBlueberry Breeding and Genomics Lab, Horticultural Sciences Department, University of Florida, Gainesville, FL, United StatesIFAS Gulf Coast Research and Education Center, University of Florida, Wimauma, FL, United StatesCentre for Research in Agricultural Genomics (CRAG) CSIC-IRTA-UAB-UB, Campus UAB, Barcelona, SpainInstitut de Recerca i Tecnologia Agroalimentàries (IRTA), Barcelona, SpainBlueberry Breeding and Genomics Lab, Horticultural Sciences Department, University of Florida, Gainesville, FL, United StatesIFAS Gulf Coast Research and Education Center, University of Florida, Wimauma, FL, United StatesCentre for Research in Agricultural Genomics (CRAG) CSIC-IRTA-UAB-UB, Campus UAB, Barcelona, SpainICREA, Passeig de Lluís Companys 23, Barcelona, SpainGenomic prediction (GP) is the procedure whereby the genetic merits of untested candidates are predicted using genome wide marker information. Although numerous examples of GP exist in plants and animals, applications to polyploid organisms are still scarce, partly due to limited genome resources and the complexity of this system. Deep learning (DL) techniques comprise a heterogeneous collection of machine learning algorithms that have excelled at many prediction tasks. A potential advantage of DL for GP over standard linear model methods is that DL can potentially take into account all genetic interactions, including dominance and epistasis, which are expected to be of special relevance in most polyploids. In this study, we evaluated the predictive accuracy of linear and DL techniques in two important small fruits or berries: strawberry and blueberry. The two datasets contained a total of 1,358 allopolyploid strawberry (2n=8x=112) and 1,802 autopolyploid blueberry (2n=4x=48) individuals, genotyped for 9,908 and 73,045 single nucleotide polymorphism (SNP) markers, respectively, and phenotyped for five agronomic traits each. DL depends on numerous parameters that influence performance and optimizing hyperparameter values can be a critical step. Here we show that interactions between hyperparameter combinations should be expected and that the number of convolutional filters and regularization in the first layers can have an important effect on model performance. In terms of genomic prediction, we did not find an advantage of DL over linear model methods, except when the epistasis component was important. Linear Bayesian models were better than convolutional neural networks for the full additive architecture, whereas the opposite was observed under strong epistasis. However, by using a parameterization capable of taking into account these non-linear effects, Bayesian linear models can match or exceed the predictive accuracy of DL. A semiautomatic implementation of the DL pipeline is available at https://github.com/lauzingaretti/deepGP/.https://www.frontiersin.org/article/10.3389/fpls.2020.00025/fullgenomic predictiongenomic selectionpolyploid speciesdeep learningepistasiscomplex traits |
| spellingShingle | Laura M. Zingaretti Salvador Alejandro Gezan Luis Felipe V. Ferrão Luis F. Osorio Amparo Monfort Amparo Monfort Patricio R. Muñoz Vance M. Whitaker Miguel Pérez-Enciso Miguel Pérez-Enciso Exploring Deep Learning for Complex Trait Genomic Prediction in Polyploid Outcrossing Species Frontiers in Plant Science genomic prediction genomic selection polyploid species deep learning epistasis complex traits |
| title | Exploring Deep Learning for Complex Trait Genomic Prediction in Polyploid Outcrossing Species |
| title_full | Exploring Deep Learning for Complex Trait Genomic Prediction in Polyploid Outcrossing Species |
| title_fullStr | Exploring Deep Learning for Complex Trait Genomic Prediction in Polyploid Outcrossing Species |
| title_full_unstemmed | Exploring Deep Learning for Complex Trait Genomic Prediction in Polyploid Outcrossing Species |
| title_short | Exploring Deep Learning for Complex Trait Genomic Prediction in Polyploid Outcrossing Species |
| title_sort | exploring deep learning for complex trait genomic prediction in polyploid outcrossing species |
| topic | genomic prediction genomic selection polyploid species deep learning epistasis complex traits |
| url | https://www.frontiersin.org/article/10.3389/fpls.2020.00025/full |
| work_keys_str_mv | AT lauramzingaretti exploringdeeplearningforcomplextraitgenomicpredictioninpolyploidoutcrossingspecies AT salvadoralejandrogezan exploringdeeplearningforcomplextraitgenomicpredictioninpolyploidoutcrossingspecies AT luisfelipevferrao exploringdeeplearningforcomplextraitgenomicpredictioninpolyploidoutcrossingspecies AT luisfosorio exploringdeeplearningforcomplextraitgenomicpredictioninpolyploidoutcrossingspecies AT amparomonfort exploringdeeplearningforcomplextraitgenomicpredictioninpolyploidoutcrossingspecies AT amparomonfort exploringdeeplearningforcomplextraitgenomicpredictioninpolyploidoutcrossingspecies AT patriciormunoz exploringdeeplearningforcomplextraitgenomicpredictioninpolyploidoutcrossingspecies AT vancemwhitaker exploringdeeplearningforcomplextraitgenomicpredictioninpolyploidoutcrossingspecies AT miguelperezenciso exploringdeeplearningforcomplextraitgenomicpredictioninpolyploidoutcrossingspecies AT miguelperezenciso exploringdeeplearningforcomplextraitgenomicpredictioninpolyploidoutcrossingspecies |