WGCNA on methylation data

Question

[enora.fremy]

Hello everyone,

I am working on methylation data and I want to use WGCNA on them. The data are the beta values of 400K CpG probes and ~30 samples which were transformed by log2(beta+1) (The data then have a nice scale-free topology profile when the soft threshold is calculated, I think it is a good sign? ). So, if someone already use WGCNA for this kind of data, I would have some questions: Firstly, are my data properly normalized? And secondly, how can I choose the maximal number of blocks for the blockWiseModule function ? I have access to a cluster, can I deduct the maxBlockSize when I know the available RAM?

Thanks in advance,

Have a nice day !

Enora

Answer 1

[Kevin Blighe]

The β (Beta) values are already normalised - I am not sure that you need to log-transform them. WGCNA is fundamentally based on correlation, so, using the β values should be fine. Can you define what you mean by a 'nice' scale-free topology? Which soft threshold did you choose?

Regarding block size, please read the WGCNA developer's response, here: https://support.bioconductor.org/p/87768/#87792

Even though my comments do not fully answer your question(s), I am providing this as an answer in the expectation that you will be able to search the various web forums / boards and, in that way, obtain more information.

Kevin

[enora.fremy]

Thank you for your comment.

I said "nice" because, the curve saturates above the 0.9 line with a soft threshold of 14 or 16. The non-transformed beta values lead to a curve as good as with the transformed data so it's good. At the end, I used the 400K beta values with a maxBlockSize of 60K (128GB available) and it worked out fine. Peter Langfelder answered me here on Bioconductor support : https://support.bioconductor.org/p/121989/#122136

Enora