Back to Home

ReMeDy: A Flexible Statistical Framework For Region-based Detection of DNA Methylation Dysregulation

Suvo Chatterjee, Siddhant Meshram, Ganesan Arunkumar, Fasil Tekola-Ayele, and Arindam Fadikar

bioRxiv preprint, 2026

Abstract

ReMeDy is a region-based statistical framework for DNA methylation studies that jointly models changes in methylation mean and variability.

Unlike common methods focused only on mean shifts, ReMeDy identifies differentially methylated, variably methylated, and joint mean-variance regions using a hierarchical likelihood GLM approach on biologically defined co-methylated regions.

The method avoids heuristic smoothing/kernel hyperparameters, controls false discovery in simulations, and improves biologically meaningful discovery in population-level studies.

Citation

BibTeX
@article{chatterjee2026remedy,
  title={ReMeDy: A Flexible Statistical Framework For Region-based Detection of DNA Methylation Dysregulation},
  author={Chatterjee, Suvo and Meshram, Siddhant and Arunkumar, Ganesan and Tekola-Ayele, Fasil and Fadikar, Arindam},
  journal={bioRxiv},
  year={2026},
  doi={10.64898/2026.01.02.697394},
  note={Preprint}
}