Fig. 1From: Classifying breast cancer subtypes on multi-omics data via sparse canonical correlation analysis and deep learningProcedure of DSCCN. Step 1 first performs differential analysis on mRNA and DNA methylation (DNAm) omics data to find DE-genes. Step 2 uses a SCCA model to detect highly correlated genes in mRNA and DNAm using DE-genes. Step 3 utilizes correlated genes to train the deep neural network model DNN to classify the breast cancer subtypesBack to article page