Building a classifier model for sub tumours?

Question

[Pin.Bioinf]

Hello,

I am wondering, If I have found that I have genes whose expression changes and CpGs whose methylation changes define sub groups of liver tumours, and I have big amounts of clinical data for patients of liver cancer,

would that data be enough to try build a model that predicts response of the patients to treatment or subgroup tumors as a function of the expression and methylation levels of my significant genes and CpGs?

How would I do this? Can you give me an example with, por example, GENE1, GENE2, GENE3 and cpg1, cpg5, cpg8, cpg9?

Thank you