Figure 1
From: Bag of Naïve Bayes: biomarker selection and classification from genome-wide SNP data
Schematics of the BoNB algorithm: B Bootstrap samples {X(1) . . . X(B)} are drawn from a GWAS training dataset X; B Naïve Bayes Classifiers (NBC) are trained on the Bootstrap samples, with the novel procedure for attribute ranking and selection; predictions of unseen subjects from a GWAS test dataset are carried out independently by each NBC and class probabilities are then averaged; biomarker selection is carried out with the novel permutation-based procedure, exploiting Out-of-Bag (OOB) samples.