What percentage change in methylation of cpg sites is biologically meaningful

Question

[confusedbiologist]

I was reading a paper and was surprised to see that many of the genes listed as significantly demethylated had ~1% differences in cpg island methylation.

can somebody suggest a paper that can explain what seems to be a small effect size is biologically meaningful

http://www.nature.com/articles/srep20544/tables/3

[igor]

I may not be fully understanding the table, but I am not sure how those can be significant. It's very rare that a change of less than 5% and especially 1% can be significant and yet there they are.

Most of the expression differences are less than 20%, which is also very dubious for microarray-based experiments.

[Devon Ryan]

Such a paper doesn't exist, because such piddly changes are biologically meaningless. You'll find a surprisingly large number of "high impact" papers making a big deal out of meaningless gene or promoter level changes, since it's easy to get a nice p-value for them.