Skip to main content
Fig. 3 | BioData Mining

Fig. 3

From: DeepAutoGlioma: a deep learning autoencoder-based multi-omics data integration and classification tools for glioma subtyping

Fig. 3

Subtype classification framework of the DeepAutoGlioma. Methylome and transcriptome data are preprocessed, differentially expressed genes (DEGs) and differentially methylated regions (DMRs) are identified, and clinically significant features are extracted. Further, these features are mapped according to the genomic region to integrate the CpG-gene pair. Then, clinically relevant methylation (CpGs) and gene expression data are fed into the autoencoder, and latent variables are extracted to build deep learning models for subtyping brain cancer

Back to article page

BioData Mining

ISSN: 1756-0381