Pan-Cancer Classification Using Multi-Omic Synergy

This project investigates the hypothesis that a machine learning model’s ability to classify cancer types can be significantly improved by integrating multi-omic data.

Cancer is a disease driven by genomic changes that manifest as functional changes in the transcriptome. While studies often use one data type, such as gene expression (RNA-seq), this project tests whether a model given both the structural blueprint (Copy Number Variation or CNV) and the functional activity (RNA-seq) will build a more robust and synergistic model, achieving a new level of predictive accuracy.

You can also view the standalone notebook here.