Tweet
Hello,
I have DNA methylation Illumina 450k data.
I processed the data in the following order:
1. convert image data to signal intense data and get row beta values.
2. the raw beta values are normalized.
3. probes are filtered twice (cross-reactive probes and commnon variant containing probes are filtered.)
4. Batch effect adjustment.
I am thinking if I should switch the step 2 and the step 3? I tend to feel that I should filter the probes before normalize the data.
Thank you very much in advance.
I have DNA methylation Illumina 450k data.
I processed the data in the following order:
1. convert image data to signal intense data and get row beta values.
2. the raw beta values are normalized.
3. probes are filtered twice (cross-reactive probes and commnon variant containing probes are filtered.)
4. Batch effect adjustment.
I am thinking if I should switch the step 2 and the step 3? I tend to feel that I should filter the probes before normalize the data.
Thank you very much in advance.