Colorectal Cancer (CRC) is an important cause of cancer-related mortality world-wide. The Consensus Molecular Subtypes represent the first comprehensive molecular classification with clinical implications, but many aspects are still missing. We use a transomic approach to improve the stratification, prognosis, and treatment prediction of CRC patients.What is the novelty of the proposed approach?Precision medicine is a reality in some tumor types, yet this cannot be said for CRC, where predictive biomarkers are scarce and are more effective at identifying non-responders than patients who may benefit from treatment. Here, we aim to provide insights into CRC and therapy responsiveness by combining different omics approaches, such as genomics, histomics, and pharmacogenomics, and integrating these by means of an AI-driven multimodal classifier. To improve the current understanding of the molecular basis of CRC we work on characterizing the relationship between morphology and molecular cancer variants in genotype, transcriptome, proteome and single-cell image data. In particular, variations in short tandem repeats of human genomes are associated with gene expression changes in samples from CRC patients. Here, we will systematically search for tandem repeat variations in tumors, annotating and cataloguing those variants that lead to tumorigenesis of colorectal cancer.