Related Experiment Video
Updated: Jun 7, 2026
14:56
Sample Preparation to Bioinformatics Analysis of DNA Methylation: Association Strategy for Obesity and Related Trait Studies
Published on: May 6, 2022
Impact of sample heterogeneity on methylation analysis.
Marie Loh1, Natalia Liem, Pei Li Lim
1Cancer Science Institute of Singapore, National University of Singapore, Singapore.
Summary
Tumor content significantly impacts DNA methylation analysis. A minimum of 70% tumor content is recommended for accurate interpretation and identification of candidate genes in gastric cancer studies.
Area of Science:
- Genomics
- Epigenetics
- Cancer Research
Background:
- High-throughput arrays are crucial for methylation analysis.
- Tumor content's influence on methylation data interpretation is not well-defined.
- Establishing acceptable tumor content thresholds is vital for reliable results.
Purpose of the Study:
- To systematically assess the impact of tumor content on methylation analysis in gastric tumors.
- To determine the optimal tumor content threshold for accurate methylation studies.
- To identify differentially methylated probes in relation to tumor purity.
Main Methods:
- Analysis of 98 gastric tumors and matched normal tissues using Illumina GoldenGate methylation assay.
- Unsupervised hierarchical clustering to group samples based on methylation profiles and tumor content.
- Differential methylation analysis comparing tumor and normal samples across varying tumor content levels.
Main Results:
- Hierarchical clustering revealed distinct sample clusters based on average tumor content.
- Differential methylation probe identification was sensitive to sample tumor content.
- Including samples with <70% tumor content reduced the sensitivity for identifying top candidate probes.
Conclusions:
- Tumor content significantly influences methylation level interpretation and candidate gene discovery.
- A tumor content threshold of 70% is suggested as suitable for selecting samples in methylation studies.
- The presence of SNPs or repetitive elements on arrays had minimal impact on findings.
