nominally significantP < 8.91 × 10 −3
When GS was left out, 11 out of the 15 tested CpG sites remained nominally significant ( P < 8.91 × 10 −3 ).
When GS was left out, 11 out of the 15 tested CpG sites remained nominally significant ( P < 8.91 × 10 −3 ).
There was an increasing trend of effect sizes as the P -value threshold for MS calculation became increasingly stringent (Fig. 3 ).