Question: Standardization of gene expression and methylation dataset to the same scale
0
gravatar for ymukhiddin
4.5 years ago by
ymukhiddin0
Czech Republic
ymukhiddin0 wrote:

I am trying to build a classification model using logistic regression with regularization based on a combined methylation and gene expression datasets. Data I am considering now is for COAD cancer from TCGA. I am going to combine selected CpG sites and gene expressions as one dataset. But, I am not sure what is the best/appropriate approach to make methylation betas and expression intensities have the same scale.

Thanks in advance for the answers.

ADD COMMENTlink modified 4.5 years ago by Sean Davis25k • written 4.5 years ago by ymukhiddin0
1
gravatar for Sean Davis
4.5 years ago by
Sean Davis25k
National Institutes of Health, Bethesda, MD
Sean Davis25k wrote:

I'd consider converting the methylation beta values to M-values and then standardizing (possibly using robust estimators such as median and IQR).

 

ADD COMMENTlink written 4.5 years ago by Sean Davis25k

Thank you. I forgot about M-vales. What do you think, if I combine M-values with gene expressions and then standardize sample-wise, for example in such a way: (feature_val - median)/IQR, would it be ok?

ADD REPLYlink written 4.5 years ago by ymukhiddin0
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 2.3.0
Traffic: 790 users visited in the last hour