Hello. How can I distinguish which genes are responsive in RNA seq data along with a continuous variable? I am not expecting a correlation between the gene count data and the variable (salinity as an example), rather I am expecting some genes to "peak' in low salinity values, other in mid-level and some at high values. I what to find the key responsive genes affected by different levels of salinity in general, and to map those associated with different salinity levels..
Thanks!
Hey there,
if you are expecting a "peak" in your data in some genes, you can try - in my opinion - two things:
1. Usual RNA-Seq analysis with like EdgeR or Deseq2 or Cuffdif (the top three for this task) or
2. You can make an easy analysis by yourself using the mean over your replicates for each gene for each condition and look at the genes with the highest alteration (using 3 standard deviations as example)
Hope this helps.
Greetings,
André
Thanks. But to make it clearer- I don't have a control, replicates, or "treatments".. This is an environmental study, so I have a large gradient of salinity values (including some zero salinity), and the "dose" is my personal interpretation.. I really don't want to devied the salinity into "low"-"mid"-"high" doses. I prefer searching for the genes responsive to salinity concentration. If I'm using EdgeR for that matter, should I insert the salinity value as a "group" factor?
Hey,
in my honest opinion, you should go for a grouping here, yes.
Greetings,
André
- Badges
- Science topic
- Similar topics
- Geoscience
- Cartography
- Mapping
- Maps