Skip to main content
Fig. 2 | BMC Bioinformatics

Fig. 2

From: HOME: a histogram based machine learning approach for effective identification of differentially methylated regions

Fig. 2

Comparison of DMR detection methods (HOME, DSS and Metilene) on simulated data. (a) Browser representation showing the quality and boundary accuracy of predicted DMRs by HOME, DSS and Metilene on two simulated classes. The horizontal bars indicate the DMRs. Simulated DMRs (black) and the scale are the same for both classes. (b) The performance of HOME, DSS, and Metilene was assessed in terms of true positive rate (TPR) and positive predictive value (PPV) for both classes. The plots show mean and standard deviation of TPR and PPV for 5 random simulations. The evaluation was performed in terms of percent reciprocal overlap ranging from 50 to 100% between simulated and predicted DMRs by HOME, DSS and Metilene for two classes

Back to article page