A review of deep learning applications for genomic selection

Abstract

Background Several conventional genomic Bayesian (or no Bayesian) prediction methods have been proposed including the standard additive genetic effect model for which the variance components are estimated with mixed model equations. In recent years, deep learning (DL) methods have been considered in the context of genomic prediction. The DL methods are nonparametric models providing flexibility to adapt to complicated associations between data and output with the ability to adapt to very complex patterns. Main body We review the applications of deep learning (DL) methods in genomic selection (GS) to obtain a meta-picture of GS performance and highlight how these tools can help solve challenging plant breeding problems. We also provide general guidance for the effective use of DL methods including the fundamentals of DL and the requirements for its appropriate use. We discuss the pros and cons of this technique compared to traditional genomic prediction approaches as well as the current trends in DL applications. Conclusions The main requirement for using DL is the quality and sufficiently large training data. Although, based on current literature GS in plant and animal breeding we did not find clear superiority of DL in terms of prediction power compared to conventional genome based prediction models. Nevertheless, there are clear evidences that DL algorithms capture nonlinear patterns more efficiently than conventional genome based. Deep learning algorithms are able to integrate data from different sources as is usually needed in GS assisted breeding and it shows the ability for improving prediction accuracy for large plant breeding data. It is important to apply DL to large training-testing data sets.

Comparison of Single-Breed and Multi-Breed Training Populations for Infrared Predictions of Novel Phenotypes in Holstein Cows

“…We performed a random grid search of hyperparameters using the h2o.grid function in the h2o R package (https://cran.r-project.org/web/packages/h2o, accessed on 7 May 2021) in order to select the optimal hyperparameters combination that minimizes the predictive loss function (i.e., prediction error) of the model for each trait (Figure 2). The random grid search was performed using the training set from each CV design (10-fold_HO, BS_HO, BS+HO_10fold, Multi-breed and Multi-breed CV2) for each trait, splitting it into a 5-fold CV [22]. Thus, 4 folds were assigned to hyperparameter optimization, aiming to find the best combination of the main hyperparameters for GBM approach, while the 1 remaining fold was used to evaluate the model performance based on the loss function (root mean square error-RMSE) and prediction accuracy (r-square-r 2 ) [19].…”

Genomic Selection in Tropical Forage Grasses: Current Status and Future Applications

“…Deep learning-based neural network studies demonstrated that accuracies must be increased by pre-trained models and data augmentation (Castro et al, 2020). Nevertheless, deep learning progress is accelerating and will be able to perform better predictions than ever (Montesinos-López et al, 2021). Although it has been the subject of debate in the past, extra investment in phenotyping technologies is becoming more accepted to capitalize on recent developments in crop genomics and prediction models.…”

Harnessing Crop Wild Diversity for Climate Change Adaptation

“…In short, ML [138,[176][177][178][179] and deep learning approaches [133,170,180,181] promise assisting the conservation [102,[182][183][184], managing [185,186], prioritization [187][188][189], and introgression [190,191] of crop wild variation from genebanks (Figure 2). ML may be particularly useful in unexplored isolated pockets of diversity, which contain allelic variants otherwise eroded from modern genotypes [192,193].…”