Búsqueda
Autor: Jaime Cuevas
Rodomiro Ortiz Jose Crossa Paulino Pérez-Rodríguez Jaime Cuevas (2021)
Potato breeding efficiency can be improved by increasing the reliability of selection and identifying promising germplasm for crossing. The data provided in these datasets were used to compare the prediction accuracy of genomic-estimated breeding values for several potato (Solanum tuberosum L.) breeding clones and released cultivars evaluated in three locations in northern and southern Sweden. The analysis included several traits such as tuber starch percentage and total tuber weight. Results of the analyses are reported in an accompanying journal article.
Dataset
Osval Antonio Montesinos-Lopez Jose Crossa Jaime Cuevas Morten Lillemo Philomin Juliana Ravi Singh (2018)
A new statistical model is presented for genomic prediction on maize and wheat data comprising multi-trait, multi-environment data.
Dataset
Johannes Martini Fernando Henrique Toledo Carolina Sansaloni Jose Crossa Jaime Cuevas Sivakumar Sukumaran (2020)
The genetic diversity housed in germplasm banks may provide valuable contributions to breeding efforts. It is important to understand the best way to introduce this diversity into elite breeding materials. This files in this dataset provide phenotypic and genotypic data used to compare genomic prediction approaches and different cross-validation scenarios on a set of wheat families obtained from crosses between elite materials and diverse germplasm bank accessions. The linked top cross population (LTP) materials analyzed in the study were screened under yield potential, drought, and heat stress conditions.
Dataset
BGGE: A new package for genomic prediction incorporating genotype by environments models
Italo Granato Jaime Cuevas Francisco Javier Luna Vázquez Jose Crossa Juan Burgueño Roberto Fritsche-Neto (2018)
One of the major issues in plant breeding is the occurrence of genotype by environment (GE) interaction. Several models have been created to understand this phenomenon and explore it by selecting the most stable genotypes. In the genomic era, several models were employed to simultaneously improve selection by using markers and account for GE interaction. Some of these models use special genetic covariance matrices. In addition, multi-environment trials scales are getting larger, and this increases the computational challenges. In this context, we propose an R package that, in general, allows building GE genomic covariance matrices and fitting linear mixed models, in particular, to a few genome GE models. Here we propose a function to create the genomic kernels needed to fit these models. This function makes genome predictions through a Bayesian linear mixed model approach. A particular treatment is given for structured dispersed covariance matrices; in particular, those structured as a block diagonal that are present in some GE models in order to decrease the computational demand. In empirical comparisons with Bayesian Genomic Linear Regression (BGLR), accuracies and the mean squared error were similar; however, the computational time was up to five times lower than when using the classic approach. Bayesian Genomic Genotype × Environment Interaction (BGGE) is a fast, efficient option to create genome GE kernels and make genomic predictions.
Dataset
Deep kernel and deep learning for genomic-based prediction
Jose Crossa Paulino Pérez-Rodríguez Juan Burgueño Ravi Singh Philomin Juliana Osval Antonio Montesinos-Lopez Jaime Cuevas (2019)
Deep learning (DL) is a promising method in the context of genomic prediction for selecting individuals early in time without measuring their phenotypes. iI this paper we compare the performance in terms of genome-based prediction of the DL method, deep kernel (arc-cosine kernel, AK) method, Gaussian kernel (GK) method and the conventional kernel method (Genomic Best Linear Unbiased Predictor, GBLUP, GB). We used two real wheat data sets for the benchmarking of these methods. We found that the GK and deep kernel AK methods outperformed the DL and the conventional GB methods, although the gain in terms of prediction performance of AK and GK was not very large but they have the advantage that no tuning parameters are required. Furthermore, although AK and GK had similar genomic-based performance, deep kernel AK is easier to implement than the GK. For this reason, our results suggest that AK is an alternative to DL models with the advantage that no tuning process is required.
Dataset