Leveraging such correlations in the predictive model might improve the prediction accuracy. Obviously some of the traits are correlated, such as weight and size. For example, for plant dataset, once we obtain a fruit, we could measure its weight, size, etc. However, in reality, it is quite often the case that we could observe and measure multiple traits rather than one, especially for crops and animals. Most of the work assumes that for each set of samples there is only one trait, and therefore, a single regression is conducted to predict the trait value.

, 2011 Park and Casella, 2008 ), as well as other machine learning methods. , 2010 ) and Bayesian Lasso ( Legarra et al. , 1998 ), Bayes A, Bayes B ( Meuwissen et al. , 2001 ), Elastic-Net, Lasso, Ridge Regression ( Tibshirani, 1994 Shaobing Chen et al. There have been lots of work on predicting genetic trait values from genotype data, such as rrBLUP (Ridge regression BLUP) ( Meuwissen et al. The Avocado data set has not been published yet and is available upon request.Д¬ontact: = ОІ 0 + в€‘ i = 1 m ОІ i X i + e l (1) where X i is the i th genotype value, m is the total number of genotypes and ОІ i is the regression coefficient for the i th genotype, e l is the error term. Our experiments show that modeling multiple traits together could improve the prediction accuracy for correlated traits.ДЄvailability and implementation: The programs we used are either public or directly from the referred authors, such as MALSAR ( ) package. We proposed a few strategies to improve the least square error of the prediction from these algorithms. We then adapted multitask learning algorithms and multiple output regression algorithms to solve the multitrait prediction problem. In this work, we view the multitrait prediction problem from a machine learning angle: as either a multitask learning problem or a multiple output regression problem, depending on whether different traits share the same genotype matrix or not. Therefore, modeling all the multiple traits together may improve the prediction accuracy. Some of these traits might be correlated with each other. In many cases, for the same set of samples and markers, multiple traits are observed.

Genetic trait prediction is usually represented as linear regression models. Given a set of biallelic molecular markers, such as SNPs, with genotype values encoded numerically on a collection of plant, animal or human samples, the goal of genetic trait prediction is to predict the quantitative trait values by simultaneously modeling all marker effects.

Antedependence strucutre asreml