Entering edit mode
Hi, I'm doing a multi-omic analysis for a Bulk RNA-seq transcriptomic, a lipidomic (from LC-MS), and a RRBS methylomic. I'm currently analyzing the lipidomic data, but confused by many packages and ways online. I wonder what R package I could use for the lipidomics ? I don't have the skyline data, so may not be able to sue 'lipidr'. I was recommended to use DESeq2, which I just need to expand the abundance by 1000000 (the abundance for each species is normalized to the total lipid intensity/abundance in the sample), but not sure what else I should modify.
There are also other questions for the lipidomics: Q:
1. Should I combine the data from +ve and -ve modes ?
2. When should I do the lipid classification, in PCA or in the visualization after differential analysis ?
3. My data is the ratio/relative abundance, so should I do any normalization before log transformation ?
4. Should I keep rare lipids (absent in many samples) ?
5. Should I keep the signals like 'TG_oxid', which maybe not the real lipids ?
6. Some signals show 1+ lipid annotation, (e.g., TG(39:01) & PE(38:02)), what should I do for them?
7. What package should I use (the main question I wonder) ?
a. Limma ?
b. DESeq2 ?
8. We don't know the number of adipocytes in each sample, just know that the same volume (50uL) of adipocytes were used in each sample when doing the experiments. The different groups' adipcyte may have different sizes, thus affecting the number of adipocytes in the same volume of sample.
a. So we measured the lipidomic data as normalized % for each sample.
b. Whether we can get 'absolute abundance' after we calculate the DNA amount in the samples (50uL), and correct the lipidomic data based on the DNA amounts ?
Thank you very much! You could answer whichever question you'd like to answer!
Code should be placed in three backticks as shown below
# include your problematic code here with any corresponding output
# please also include the results of running the following in an R session
sessionInfo( )
