Prediction of the importance of auxiliary traits using computational intelligence and machine learning: A simulation study

Abstract

The present study evaluated the importance of auxiliary traits of a principal trait based on phenotypic information and previously known genetic structure using computational intelligence and machine learning to develop predictive tools for plant breeding. Data of an F2 population represented by 500 individuals, obtained from a cross between contrasting homozygous parents, were simulated. Phenotypic traits were simulated based on previously established means and heritability estimates (30%, 50%, and 80%); traits were distributed in a genome with 10 linkage groups, considering two alleles per marker. Four different scenarios were considered. For the principal trait, heritability was 50%, and 40 control loci were distributed in five linkage groups. Another phenotypic control trait with the same complexity as the principal trait but without any genetic relationship with it and without pleiotropy or a factorial link between the control loci for both traits was simulated. These traits shared a large number of control loci with the principal trait, but could be distinguished by the differential action of the environment on them, as reflected in heritability estimates (30%, 50%, and 80%). The coefficient of determination were considered to evaluate the proposed methodologies. Multiple regression, computational intelligence, and machine learning were used to predict the importance of the tested traits. Computational intelligence and machine learning were superior in extracting nonlinear information from model inputs and quantifying the relative contributions of phenotypic traits. The R2 values ranged from 44.0% - 83.0% and 79.0% - 94.0%, for computational intelligence and machine learning, respectively. In conclusion, the relative contributions of auxiliary traits in different scenarios in plant breeding programs can be efficiently predicted using computational intelligence and machine learning.

Computational Intelligence to Study the Importance of Predictors in White Oat ( <i>Avena Sativa</i> L.)

“…Stepwise multiple regression is the variable selection method, which aims to explain the relationship between a set of independent variables and a dependent variable. The coe cient of determination (R 2 ) aims to estimate how much of the independent variable is explained by the total variation of the dependent variable [3,4].…”

“…The importance of predictors through the PMC network was quanti ed using two techniques. The rst, based on Garson's (1991) [7] algorithm modi ed by Goh (1995)[6], consists of partitioning the neural network connection weights to determine the relative importance of each input variable within the network [3,4].…”

“…After the network is established, the importance of variables (inputs) can also be obtained, considering the impact of destructuring or disturbing the information of a given input on the estimation of the coe cient of determination [3,4].…”

“…Estimating the importance of predictor variables in breeding programs allows for faster progress, selecting, and predicting traits with low heritability and/or measurement di culty [3,4]. Although simultaneous assessment of characteristics provides a wide variety of information, identifying which predictor variable is more critical is challenging for the breeder [5].…”

“…[4] used these methodologies to predict grain yield, grain length-width ratio, and panicle length in ood-irrigated rice. [3] evaluated the importance of auxiliary traits of the main trait based on phenotypic information and previously known genetic structure using computational intelligence and machine learning to develop good predictive tools in breeding programs. However, there are no studies in the literature related to yield prediction and veri cation of the importance of variables for grain yield in white oat culture.…”