Fig. 2: Genomic prediction accuracy of three levels of selected SNPs (1, 10, and 50% of all markers) using Bayes-A, Bayes-B, Bayes-C, Bayes LASSO (BL), and Bayesian ridge regression (BRR) models and different approaches to SNP selection. | Heredity

Fig. 2: Genomic prediction accuracy of three levels of selected SNPs (1, 10, and 50% of all markers) using Bayes-A, Bayes-B, Bayes-C, Bayes LASSO (BL), and Bayesian ridge regression (BRR) models and different approaches to SNP selection.

From: Integration of single nucleotide variants and whole-genome DNA methylation profiles for classification of rheumatoid arthritis cases from controls

Fig. 2

Randomly (a), evenly spaced (b), GWAS p value (c), and strongest BRR effects (d). AUCs with different superscript(s) are significantly different from each other (p value < 0.05).

Back to article page