Skip to main content
Fig. 4 | BMC Bioinformatics

Fig. 4

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

Fig. 4

RNA transcript sequence contains relevant information for ASD risk gene prediction. a Model performance with different k-mer combinations (k = 1, 2, 3 or 4). Based on the ROC AUC, 3-mers and 4-mers were selected as the sequence features. b Model performance with different numbers of selected k-mers. The 3-mers and 4-mers were ranked according to the importance scores provided by the RF model, and the top 25 k-mers were selected for model construction

Back to article page