Skip to main content
Fig. 2 | BMC Bioinformatics

Fig. 2

From: A semi-supervised deep learning approach for predicting the functional effects of genomic non-coding variations

Fig. 2

Feature distribution and the predictive performance of our deep learning model. a Plots showing the distribution of each score in the input feature map after preprocessing for K562. b Pearson correlation of each feature vector with the labels of non-coding variants in K562. c ROC curves showing the performance of our model. d AUC values showing the performance of our model with each of the grouped features in GM12878. TPR true positive rate, FDR false discovery rate, AUC area under the ROC (receiver operating characteristic) curve, group I histone marks on enhancers, group II histone marks on promoters, group III structural histone marks, group IV heterochromatin histone marks, group V histone marks on transcribed gene-body, group VI DNA accessibility assayed by DNase I sensitivity

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us