The Madzo lab focuses on computational biology, particularly on understanding DNA methylation drift during normal aging, disease-driven inflammation, and neoplastic transformation. Previous work has shown that cells exhibit changes in DNA methylation patterns during aging. Notably, diet and calorie restriction have been found to slow down age- or inflammation-related DNA methylation drift. Interestingly, cells undergoing malignant transformation display similar DNA methylation drift, suggesting that cancer cells exhibit a phenotype akin to accelerated aging.

We are particularly focused on utilizing epigenetic drugs, such as inhibitors of DNA methyltransferases, to reactivate silenced repetitive elements in cancer cells. Reactivation of these elements leads to an inflammation-like phenotype, making the cancer cells more immunogenic and sensitized to immunotherapy.

We are working on multiple projects related to epigenetic changes during aging and cancer development. Our projects involve the integrated analysis of multiple sequencing datasets, including RNAseq, ATACseq, and scRNA-seq, to uncover the mechanisms involved in the genetic and epigenetic regulation of repetitive DNA.

A substantial part of our work involves querying high-dimensional data from publicly available sources such as TCGA, ENCODE, and GEO portals. We utilize existing open-source bioinformatics tools along with custom Python and R scripts. In our data analysis, we apply statistical methods such as permutation, multivariate analysis, and machine learning techniques like random forest, XGBoost, and stochastic gradient descent.