Regression model for chronological age based on epigenetic data

In this scientific project, I built a new model to predict the chronological age of subjects using whole blood DNA methylation data.

I mined and processed data from public repositories to build large, heterogenous datasets for model training, testing and validation.

Using Python and the scikit-learn library, I built regularized regression models which ended up performing on par with state-of-the-art approaches.

You can read the whole story on the pre-print linked below.