Hi Seniors and all members,
Hope you all are well. My question may be answered somewhere in Biostar, but after reading many posts, it seems to me that it did not solve the problem I am dealing with. My big apology for asking and understand you all are busy, but if you have a moment to suggest, I am really thankful for that.
I have bulk RNAseq data from 24 tumour samples from mice, which splitted into six groups (4 per group) and each received different dose of radiation treatment (Untreated (0Gy), 2Gy_1 fraction, 2Gy_2 fractions....etc). so far, I have done differentially expressed genes analysis and some other pathways, enrichment...etc.
I am interested in doing WGCNA for gene that are co-expressed. I followed the WGCNA tutorial and been able to identify the different modules. Therefore, the next step is to find out the important module for further analysis.
I do not have any traits besides the different treatment conditions, which are different doses of radiation treatment. I considered treatment as the categorical variables (6 groups). Basically, correlating module to treatment conditions is not possible because one is numeric and another one is categorical variable with many levels. Stardard logistic regression takes only binary outcome.
what is normally done to associate WGCNA modules with different level of categorical traits variables ( in my case many level of radiation treatments).
Thank you very much beforehand for taking time to read the questions.
Regards,
synat
