Abstract
Many methodologies are used to predict the genetic merit in animals and plants, but some of them require priori assumptions that may increase the complexity of the model. Artificial neural network (ANN) has advantage to not require priori assumptions about the relationships between inputs and the output allowing great flexibility to handle different types of complex non-additive effects, such as dominance and epistasis. Despite this advantage, the biological interpretability of ANNs is still limited. The aim of this research was to estimate the heritability and markers effects for two traits in Coffea canephora using an additive-dominance architecture ANN and to compare it with genomic best linear unbiased prediction (GBLUP). The data used consists of 51 clones of C. canephora varietal Conilon, 32 of varietal group Robusta and 82 intervarietal hybrids. From this, 165 phenotyped individuals were genotyped for 14,387 SNPs. Due to the high computational cost of ANNs, we used Bagging decision tree to reduce the dimensionality of the data, selecting the markers that accumulated 70% of the total importance. An ANN with three hidden layers was run, each varying from 1 to 40 neurons summing 64,000 neural networks. The network architectures with the best predictive ability were selected. The best architectures were composed by 4, 15, and 33 neurons in the first, second and third hidden layers, respectively, for yield, and by 13, 20, and 24 neurons, respectively for rust resistance. The predictive ability was greater when using ANN with three hidden layers than using one hidden layer and GBLUP, with 0.72 and 0.88 for yield and coffee leaf rust resistance, respectively. The concordance rate (CR) of the 10% larger markers effects among the methods varied between 10% and 13.8%, for additive effects and between 5.4% and 11.9% for dominance effects. The narrow-sense () and dominance-only () heritability estimates were 0.25 and 0.06, respectively, for yield, and 0.67 and 0.03, respectively for rust resistance. The ANN was able to estimate the heritabilities from an additive-dominance genomic architectures and the ANN with three hidden layers obtained best predictive ability when compared with those obtained from GBLUP and ANN with one hidden layer.
Highlights
The interest in semi- and non-parametric statistical methods for genome-enabled prediction is increasing [1]
Heritabilities from an additive-dominance genomic architectures and the Artificial Neural Networks (ANN) with three hidden layers obtained best predictive ability when compared with those obtained from genomic best linear unbiased prediction (GBLUP) and ANN with one hidden layer
64,000 neural networks were performed, with each hidden layer ranging from 1 to 40 neurons, and the ANN was chosen based on the best predictive ability
Summary
The interest in semi- and non-parametric statistical methods for genome-enabled prediction is increasing [1]. ANN is a methodology inspired by the biological behavior of human brain. ANN comprises layers divided into units called neurons. Each neuron’s output is expressed as the sum of inputs to a neuron, regulating specific weights for the predictor variables through linear and nonlinear activation functions [1, 6]. ANN have been applied for genomic prediction of complex traits in some crops as maize, eucalypt [7], soybean [8] and wheat [9]. This approach does not require making a priori assumptions about the relationships between inputs (SNP markers) and the output (phenotypic observations). The non-priori assumptions allow for great flexibility to handle different types of complex non-additive effects, such as dominance and epistasis [1, 10, 11]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
