search

yiluyucheng / dnaMethyAge

Predict epigenetic age from DNA methylation data
34 stars 10 forks source link

PCGrimAge with more CpGs than the original clock? #7

Open thscandolara opened 9 months ago

thscandolara commented 9 months ago

Hello!

I have a question regarding PCGrimAge clock CpG's number. Why is it using 78465 CpG's if the original clock is constructed with 1030 CpG's? I have even checked the version 2 of the clock (https://www.aging-us.com/article/204434/text), but they did not changed the number of CpGs used. "We used the same set of 1030 CpGs to construct version 2 of GrimAge."

Am I missing something? Thanks!!

yiluyucheng commented 9 months ago

The coefficients of the original GrimAge clock were not publicly available. Here I implemented its PC version (published in Nature Aging). The PC-based authors claimed "the PC clocks demonstrated equivalent or improved prediction of mortality and similar associations with a wide range of other factors".

And the PCGrimAge requires 78464 CpG as inputs.

thscandolara commented 9 months ago

Oh, I see. Thank you!