Deep neural networks for genomic prediction do not estimate marker effects

Abstract Genomic prediction is a promising technology for advancing both plant and animal breeding, with many different prediction models evaluated in the literature. It has been suggested that the ability of powerful nonlinear models, such as deep neural networks, to capture complex epistatic effec...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Jordan Ubbens, Isobel Parkin, Christina Eynck, Ian Stavness, Andrew G. Sharpe
Formato: article
Lenguaje:EN
Publicado: Wiley 2021
Materias:
Acceso en línea:https://doaj.org/article/33a2fce533d7499b90a08552cbaacd06
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:33a2fce533d7499b90a08552cbaacd06
record_format dspace
spelling oai:doaj.org-article:33a2fce533d7499b90a08552cbaacd062021-12-05T07:50:11ZDeep neural networks for genomic prediction do not estimate marker effects1940-337210.1002/tpg2.20147https://doaj.org/article/33a2fce533d7499b90a08552cbaacd062021-11-01T00:00:00Zhttps://doi.org/10.1002/tpg2.20147https://doaj.org/toc/1940-3372Abstract Genomic prediction is a promising technology for advancing both plant and animal breeding, with many different prediction models evaluated in the literature. It has been suggested that the ability of powerful nonlinear models, such as deep neural networks, to capture complex epistatic effects between markers offers advantages for genomic prediction. However, these methods tend not to outperform classical linear methods, leaving it an open question why this capacity to model nonlinear effects does not seem to result in better predictive capability. In this work, we propose the theory that, because of a previously described principle called shortcut learning, deep neural networks tend to base their predictions on overall genetic relatedness rather than on the effects of particular markers such as epistatic effects. Using several datasets of crop plants [lentil (Lens culinaris Medik.), wheat (Triticum aestivum L.), and Brassica carinata A. Braun], we demonstrate the network's indifference to the values of the markers by showing that the same network, provided with only the locations of matches between markers for two individuals, is able to perform prediction to the same level of accuracy.Jordan UbbensIsobel ParkinChristina EynckIan StavnessAndrew G. SharpeWileyarticlePlant cultureSB1-1110GeneticsQH426-470ENThe Plant Genome, Vol 14, Iss 3, Pp n/a-n/a (2021)
institution DOAJ
collection DOAJ
language EN
topic Plant culture
SB1-1110
Genetics
QH426-470
spellingShingle Plant culture
SB1-1110
Genetics
QH426-470
Jordan Ubbens
Isobel Parkin
Christina Eynck
Ian Stavness
Andrew G. Sharpe
Deep neural networks for genomic prediction do not estimate marker effects
description Abstract Genomic prediction is a promising technology for advancing both plant and animal breeding, with many different prediction models evaluated in the literature. It has been suggested that the ability of powerful nonlinear models, such as deep neural networks, to capture complex epistatic effects between markers offers advantages for genomic prediction. However, these methods tend not to outperform classical linear methods, leaving it an open question why this capacity to model nonlinear effects does not seem to result in better predictive capability. In this work, we propose the theory that, because of a previously described principle called shortcut learning, deep neural networks tend to base their predictions on overall genetic relatedness rather than on the effects of particular markers such as epistatic effects. Using several datasets of crop plants [lentil (Lens culinaris Medik.), wheat (Triticum aestivum L.), and Brassica carinata A. Braun], we demonstrate the network's indifference to the values of the markers by showing that the same network, provided with only the locations of matches between markers for two individuals, is able to perform prediction to the same level of accuracy.
format article
author Jordan Ubbens
Isobel Parkin
Christina Eynck
Ian Stavness
Andrew G. Sharpe
author_facet Jordan Ubbens
Isobel Parkin
Christina Eynck
Ian Stavness
Andrew G. Sharpe
author_sort Jordan Ubbens
title Deep neural networks for genomic prediction do not estimate marker effects
title_short Deep neural networks for genomic prediction do not estimate marker effects
title_full Deep neural networks for genomic prediction do not estimate marker effects
title_fullStr Deep neural networks for genomic prediction do not estimate marker effects
title_full_unstemmed Deep neural networks for genomic prediction do not estimate marker effects
title_sort deep neural networks for genomic prediction do not estimate marker effects
publisher Wiley
publishDate 2021
url https://doaj.org/article/33a2fce533d7499b90a08552cbaacd06
work_keys_str_mv AT jordanubbens deepneuralnetworksforgenomicpredictiondonotestimatemarkereffects
AT isobelparkin deepneuralnetworksforgenomicpredictiondonotestimatemarkereffects
AT christinaeynck deepneuralnetworksforgenomicpredictiondonotestimatemarkereffects
AT ianstavness deepneuralnetworksforgenomicpredictiondonotestimatemarkereffects
AT andrewgsharpe deepneuralnetworksforgenomicpredictiondonotestimatemarkereffects
_version_ 1718372546923986944