Skip to main content
Fig. 3 | BMC Bioinformatics

Fig. 3

From: Prediction and prioritization of autism-associated long non-coding RNAs using gene expression and sequence features

Fig. 3

Model performance with unsupervised representation learning of developmental brain gene expression data. a Selection of the code dimension size for the autoencoder network. For each dimension size, models were evaluated using the average performance value from fifty repetitions of tenfold cross-validations. Based on the ROC AUC, 48 was selected as the code dimension size for the autoencoder network. b ROC curves to compare the models trained using the full set of 524 expression features and the 48 encoded features

Back to article page