AUTHOR=Montesinos-López Osval A. , Montesinos-López Abelardo , Mosqueda-González Brandon A. , Bentley Alison R. , Lillemo Morten , Varshney Rajeev K. , Crossa José 

TITLE=A New Deep Learning Calibration Method Enhances Genome-Based Prediction of Continuous Crop Traits

JOURNAL=Frontiers in Genetics

VOLUME=Volume 12 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2021.798840

DOI=10.3389/fgene.2021.798840

ISSN=1664-8021

ABSTRACT=Genomic selection (GS) has the potential to revolutionize predictive plant breeding.   A reference population is phenotyped and genotyped and used to train a statistical model that is used to perform genome-enabled predictions of new individuals that were only genotyped. For this reason, deep neural networks, a form of machine learning model, have been adopted for use in GS. However, the training process for deep neural networks is very challenging due to the numerous hyper-parameters that need to be tuned. Imperfect tuning can result in biased predictions. In this paper we propose a simple method for calibrating (adjusting) continuous predictions resulting from deep learning applications. We evaluated the proposed deep learning calibration method (DL_M2) using four crop breeding data sets and its performance was compared with the standard deep learning method (DL_M1) and with the standard genomic Best Linear Unbiased Predictor (GBLUP). Overall, the GBLUP was the most accurate model but the proposed deep learning calibration method (DL_M2) helped to increase the genome-enabled prediction performance in all data sets as compared with the traditional DL method (DL_M1). Taken together we provide evidence for extending the use of the proposed calibration method in order to evaluate its potential and consistency for predicting performance in the context of GS applied to plant breeding.