Related Experiment Video
Updated: Jun 13, 2026
12:34
DNA Methylation: Bisulphite Modification and Analysis
Published on: October 21, 2011
106.7K
A comparison of differential DNA methylation analysis methods for continuous outcomes: implications for epigenetic
Oladejo Ahmodu1, Gaurav Bhatti1, Adi L Tarca1,2,3
1Center for Molecular Medicine and Genetics, Wayne State University, Detroit, MI, USA.
Epigenomics
|March 19, 2026
Summary
Comparing analysis methods for epigenome-wide association studies reveals limma offers superior reproducibility and predictive accuracy for identifying differentially methylated CpGs associated with age.
Area of Science:
- Epigenetics and Genomics
- Computational Biology
- Statistical Genetics
Background:
- Univariate methods are standard in epigenome-wide association studies (EWAS) for linking CpGs to traits.
- The performance of these common univariate methods in EWAS requires thorough evaluation.
Purpose of the Study:
- To compare the performance of limma, Spearman's correlation (SC), and quantile regression (QR) for EWAS.
- To assess methods based on reproducibility, genomic distribution, and predictive accuracy of identified CpGs.
Main Methods:
- Comparative analysis of limma, Spearman's correlation, and quantile regression on methylation data.
- Evaluation metrics included CpG reproducibility, genomic location clustering, and predictive modeling accuracy (random forest, elastic net).
Main Results:
- Limma identified more consistent gestational age-associated CpGs than SC and QR.
- CpGs identified by limma and SC showed greater chromosomal clustering compared to QR.
- Predictive models for age using top CpGs showed higher accuracy with limma and SC.
Conclusions:
- The choice of differential methylation analysis method significantly impacts CpG reproducibility, co-location, and predictive performance.
- Limma demonstrates a robust balance of reproducibility and predictive value for age-related EWAS.
