I did differential analysis on samples treated (G1E-ER4) with estradiol (differentiation induced by estradiol) and control (G1E). The genes that I gor as differentially expressed in G1E-ER4 showed following pathways.

I can explain the first five: estradiol causes stress and more hemoglobin is synthesized during cell differentiation, also tetrypyrrole is involved in hemoglobin synthesis. Cell death might also be caused by stress induced by estradiol.

However, I am not sure why I have cell division or starvation etc pathways.What is considered to be a starvation actually?

response to stress (GO:0006950)
heme metabolic process (GO:0042168)
tetrapyrrole metabolic process (GO:0033013)
regulation of apoptotic process (GO:0042981)
cell death (GO:0008219), death (GO:0016265),programmed cell death (GO:0012501)

mitotic cell cycle (GO:0000278), cell cycle (GO:0007049)
cell division (GO:0051301),nuclear division (GO:0000280)
cellular response to starvation (GO:0009267)
positive regulation of GTPase activity (GO:0043547)
regulation of localization (GO:0032879)
  • $\begingroup$ Maybe some mTOR crosstalk with your estradiol? As for your starvation and look here, it's hard to tell unless we know what kind of starvation it was, but I suspect it has something to do with the cell division. $\endgroup$ – CKM Feb 3 '16 at 18:32

Pathway analysis can be a little tricky. Because pathway enrichment is estimated from the enrichment of genes (or miRNA or what have you) that participate in overlapping pathways, certain pathways often show up for reasons that may have little to do with your experiment. Cell cycle related pathways, in particular, come up a lot, since so many genes are "cell cycle-related". These papers both go over this phenomenon in more detail. The second one is behind a paywall...sorry for that. If you can't access it, but want to read it, I'm sure we can figure out a way to send a pdf.

I deal with this a lot in my own research. My usual strategy is to make pathway analysis only one part of the overall analysis. How that works exactly depends a lot on the nature of your project. For instance, pathway enrichment could just serve as a sanity check, to see that other results aren't widely off mark. Alternatively, you might want to put other results in the context of some pathway or the other & having pathway enrichment data would naturally be a part of this.

Regarding the question of what exactly starvation means in the context of your results, it's possible that, if this pathway doesn't make sense in the context of your experiment, then it shows up as enriched because more starvation-related genes came up as differentially expressed. The degree to which that means that your cells were undergoing starvation could be debatable. I would bet money that there's significant overlap between starvation-related gene sets and those related to "cellular stress" & "cell death". Cell death is one common response to serum starvation, for instance.

Hope that helps!

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.