Question: Taking relatedness among individuals into account in differential methylation analysis?
0
gravatar for elgomez
11 months ago by
elgomez0
elgomez0 wrote:

Hi all,

I am looking for differentially methylated probes and regions with age in a sample set that includes related individuals.

So far I have been using minfi and ChAMP to analyze my 450K and EPIC array data. I've really enjoyed using both of these excellent packages and am grateful to their developers!

But I am wondering if anyone knows of a way to incorporate relatedness (for example, a relatedness matrix) into the linear model used to identify DMPs or into the bumphunter analysis?

Thank you for any thoughts on this,

EG 

minfi champ methylation • 260 views
ADD COMMENTlink modified 10 months ago by Kasper Daniel Hansen6.4k • written 11 months ago by elgomez0

Um...Sorry that I think ChAMP cannot incorporate relatedness matrix, I have never done project involved with that, thus I did not develop this functionality in ChAMP.

Maybe you can consider minfi?

Best

Yuan Tian 

ADD REPLYlink written 11 months ago by Yuan Tian110

Thank you for the reply and suggestion.

 

ADD REPLYlink written 11 months ago by elgomez0
Answer: Taking relatedness among individuals into account in differential methylation an
0
gravatar for Kasper Daniel Hansen
10 months ago by
United States
Kasper Daniel Hansen6.4k wrote:
At first, I would use "standard" tools from GWAS to do this (not saying this is easy), either on data in the unit interval or the logit transformation. In principle all you need is a mixed effect model where the covariance matrix reflects the relatedness structure. On Sat, Aug 25, 2018 at 12:34 AM eeheberlein [bioc] < noreply@bioconductor.org> wrote: > Activity on a post you are following on support.bioconductor.org > > User eeheberlein <https: support.bioconductor.org="" u="" 17056=""/> wrote Question: > Taking relatedness among individuals into account in differential > methylation analysis? <https: support.bioconductor.org="" p="" 112413=""/>: > > Hi all, > > I am looking for differentially methylated probes and regions with age in > a sample set that includes related individuals. > > So far I have been using minfi and ChAMP to analyze my 450K and EPIC array > data. I've really enjoyed using both of these excellent packages and am > grateful to their developers! > > But I am wondering if anyone knows of a way to incorporate relatedness > (for example, a relatedness matrix) into the linear model used to identify > DMPs or into the bumphunter analysis? > > Thank you for any thoughts on this, > > EG > ------------------------------ > > Post tags: methylation, minfi, ChAMP > > You may reply via email or visit > Taking relatedness among individuals into account in differential methylation analysis? >
ADD COMMENTlink written 10 months ago by Kasper Daniel Hansen6.4k

Thanks for pointing me in the right direction. I had been looking into GEMMA or Fast-LMM to do something like this, but was getting stuck on how to run programs meant to take genotype data as input on continuous methylation count or beta value data. That is where the unit interval or logit transformation comes in?

ADD REPLYlink written 10 months ago by elgomez0
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 16.09
Traffic: 364 users visited in the last hour