Related Experiment Video
Updated: Jan 3, 2026
09:45
Detection of Copy Number Alterations Using Single Cell Sequencing
Published on: February 17, 2017
12.0K
Critical evaluation of copy number variant calling methods using DNA methylation
Varun Kilaru1, Anna K Knight1, Seyma Katrinli1
1Department of Gynecology and Obstetrics, Emory University School of Medicine, Atlanta, Georgia.
Genetic Epidemiology
|November 19, 2019
Summary
Copy number variants (CNVs) can be identified using DNA methylation data. This study assessed the reliability of three methods (ChAMP, Conumee, cnAnalysis450k) for CNV calling, finding low overall reliability but improved performance with the MethylationEPIC array.
Area of Science:
- Genomics
- Epigenetics
- Bioinformatics
Background:
- Array-based DNA methylation data is increasingly used for copy number variant (CNV) detection.
- ChAMP, Conumee, and cnAnalysis450k are popular computational methods for CNV calling from methylation data.
- The reliability of these methods using real-world samples has not been previously evaluated.
Purpose of the Study:
- To assess the consistency and reliability of CNV calls derived from DNA methylation data compared to genotype data.
- To evaluate the performance of ChAMP, Conumee, and cnAnalysis450k methods across two methylation arrays: HumanMethylation450 and MethylationEPIC.
- To determine the impact of array type on CNV calling reliability.
Main Methods:
- Utilized a cohort with both genotype and DNA methylation data from HumanMethylation450 and MethylationEPIC BeadChips.
- Assessed CNV call concordance between methylation-derived and genotype-derived calls.
- Compared the reliability of CNV calls generated by ChAMP, Conumee, and cnAnalysis450k using repeated methylation measures.
Main Results:
- ChAMP identified more CNVs than Conumee and cnAnalysis450k, showing higher overlap with genotype data (~62%).
- All tested methods demonstrated relatively low reliability.
- Conumee achieved the highest reliability (57.6%) on the MethylationEPIC array, while cnAnalysis450k showed the highest reliability (43.0%) on the HumanMethylation450 array.
- The MethylationEPIC array offered improved CNV calling reliability compared to the HumanMethylation450 array, though overlap with genotype data did not significantly increase.
Conclusions:
- While DNA methylation arrays can detect CNVs, current computational methods exhibit limited reliability.
- The MethylationEPIC array provides enhanced reliability for CNV calling compared to the HumanMethylation450 array.
- Further methodological development is needed to improve the accuracy and consistency of CNV detection from methylation data.
