Skip to main content
Fig. 2 | BMC Bioinformatics

Fig. 2

From: A convolutional neural network-based regression model to infer the epigenetic crosstalk responsible for CG methylation patterns

Fig. 2

Application of epiNet models to predict the CG methylation pattern of Setd2 KO FGOs. a Pearson correlation coefficients between the predicted and actual CG methylation patterns of Setd2 KO FGOs. epiNet models trained with the data of the indicated feature(s) from wildtype FGOs were used to predict the CG methylation pattern for the input data of the same feature(s) from Setd2 KO FGOs. The Setd2 KO data were from the entire genome. b A representative genome browser shot showing the predicted CG methylation patterns of Setd2 KO FGOs. The actual CG methylation patterns of wildtype and Setd2 KO FGOs are shown for comparison. Genomic regions showing CG methylation gains are highlighted in yellow. RefSeq genes are indicated at the bottom. c The effect of additional features on the prediction of the CG methylation pattern of Setd2 KO FGOs. The details are the same as in a

Back to article page