phenotypes and kernels

The G3_R1.RData file contains the necessary data to replicate the main results found in the paper titled: 

(Quasi) Multi-task support vector regression with heuristic hyperparameter optimization for whole-genome prediction of complex traits: a case study with carcass traits in broilers

Alves, AAC, Fernandes AFA, Lopes FB, Breen V, Hawken R, Gianola D, Rosa GJM

Data Structure
The file contains the following objects:   

pheno: A data frame containing the renumbered ID for the animals, the information about training, validation and testing sets, the de-regressed proofs for CT1, Growth, CT2, and FE, scaled to lie between 0 and 1.  

D: An NxN Euclidean Distance Matrix (EDM) for the animals' genotypes.  

G: The NxN genomic relationship matrix.  

WD_CT1: A list with 3 EDMs weighted according to the genetic architecture of CT1 (D11), CT1 x Growth (D12) and Growth (D22).  

WD_CT2: A list with 3 EDMs weighted according to the genetic architecture of CT2 (D11), CT2 x FE  (D12) and FE (D22).   

Disclaimer
This dataset is for academic use only