Limits...
Power and sample size estimation for epigenome-wide association scans to detect differential DNA methylation.

Tsai PC, Bell JT - Int J Epidemiol (2015)

Bottom Line: We performed simulations to estimate power under the case-control and discordant MZ twin EWAS study designs, under a range of epigenetic risk effect sizes and conditions.Our analyses highlighted several factors that significantly influenced EWAS power, including sample size, epigenetic risk effect size, the variance of DNA methylation at the locus of interest and the correlation in DNA methylation patterns within the twin sample.Our results can help guide EWAS experimental design and interpretation for future epigenetic studies.

View Article: PubMed Central - PubMed

Affiliation: Department of Twin Research and Genetic Epidemiology, King's College London, London, UK.

No MeSH data available.


DNA methylation variance and correlation can impact EWAS power. Case-control power estimates (A, upper panel) are shown under different pooled SDs and methORs at a fixed mean difference = 10% using parametric (left panel) and nonparametric (right panel) test statistics. MZ twin power estimates (B, lower panel) are shown under different pooled SDs and correlation coefficients at a fixed mean difference = 9% using parametric (left panel) and nonparametric (right panel) test statistics.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4588864&req=5

dyv041-F5: DNA methylation variance and correlation can impact EWAS power. Case-control power estimates (A, upper panel) are shown under different pooled SDs and methORs at a fixed mean difference = 10% using parametric (left panel) and nonparametric (right panel) test statistics. MZ twin power estimates (B, lower panel) are shown under different pooled SDs and correlation coefficients at a fixed mean difference = 9% using parametric (left panel) and nonparametric (right panel) test statistics.

Mentions: We explored the effect of DNA methylation variance on EWAS power by estimating the pooled SD in DNA methylation for the combined case-control sample as a measure of variance (Figure 5A). We selected permutations with 20 cases and 20 controls at a 10% methylation mean difference and with equal variances, and estimated power by categorizing the pooled SD into four groups (0.145–0.150, 0.150–0.155, 0.155–0.160 and 0.160–0.165) and methOR into six groups (1.62–1.64, 1.64–1.66, 1.66–1.68, 1.68–1.70, 1.70–1.72 and 1.72–1.74). Power was estimated using the t test (Figure 5A, left panel) and Wilcoxon test (Figure 5A, right panel) at nominal significance. Under the t test, the pooled SD greatly influences power where greater pooled SD will lead to lower power irrespective of methOR differences. In comparison, both pooled SD and methOR have an influence on power estimated using the Wilcoxon test. Greatest power can be achieved with smaller pooled SD and at highest methOR.Figure 5.


Power and sample size estimation for epigenome-wide association scans to detect differential DNA methylation.

Tsai PC, Bell JT - Int J Epidemiol (2015)

DNA methylation variance and correlation can impact EWAS power. Case-control power estimates (A, upper panel) are shown under different pooled SDs and methORs at a fixed mean difference = 10% using parametric (left panel) and nonparametric (right panel) test statistics. MZ twin power estimates (B, lower panel) are shown under different pooled SDs and correlation coefficients at a fixed mean difference = 9% using parametric (left panel) and nonparametric (right panel) test statistics.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4588864&req=5

dyv041-F5: DNA methylation variance and correlation can impact EWAS power. Case-control power estimates (A, upper panel) are shown under different pooled SDs and methORs at a fixed mean difference = 10% using parametric (left panel) and nonparametric (right panel) test statistics. MZ twin power estimates (B, lower panel) are shown under different pooled SDs and correlation coefficients at a fixed mean difference = 9% using parametric (left panel) and nonparametric (right panel) test statistics.
Mentions: We explored the effect of DNA methylation variance on EWAS power by estimating the pooled SD in DNA methylation for the combined case-control sample as a measure of variance (Figure 5A). We selected permutations with 20 cases and 20 controls at a 10% methylation mean difference and with equal variances, and estimated power by categorizing the pooled SD into four groups (0.145–0.150, 0.150–0.155, 0.155–0.160 and 0.160–0.165) and methOR into six groups (1.62–1.64, 1.64–1.66, 1.66–1.68, 1.68–1.70, 1.70–1.72 and 1.72–1.74). Power was estimated using the t test (Figure 5A, left panel) and Wilcoxon test (Figure 5A, right panel) at nominal significance. Under the t test, the pooled SD greatly influences power where greater pooled SD will lead to lower power irrespective of methOR differences. In comparison, both pooled SD and methOR have an influence on power estimated using the Wilcoxon test. Greatest power can be achieved with smaller pooled SD and at highest methOR.Figure 5.

Bottom Line: We performed simulations to estimate power under the case-control and discordant MZ twin EWAS study designs, under a range of epigenetic risk effect sizes and conditions.Our analyses highlighted several factors that significantly influenced EWAS power, including sample size, epigenetic risk effect size, the variance of DNA methylation at the locus of interest and the correlation in DNA methylation patterns within the twin sample.Our results can help guide EWAS experimental design and interpretation for future epigenetic studies.

View Article: PubMed Central - PubMed

Affiliation: Department of Twin Research and Genetic Epidemiology, King's College London, London, UK.

No MeSH data available.