Entering edit mode
8.3 years ago
Learner
▴
280
I have a set of genes and I calculated fold change and based on that by IPA I found several pathways which are activated. however, now, I am wondering what to show in the paper ? I would appreciate any comment
Anything you see in IPA should be exportable as tables/images etc.
@genomax2 yes, I do know! but for example it says those that have fold change positive are activated which is logical. but my question is that how do you present activation in a paper ? do you have any example ? do you have any suggestion ? especially if the pathway is very large and complicated to show
IPA gives you some values that show the extent and significance of activation. Why not use those?
This is where importing in your own expression/other values for genes may help color those pathway plots (up/down).
@igor do you mean the fold change? or Z-socre? or -log(p-value) or ratio? and also do you just mention that in a paper ? is it sufficient ?
@genomax2 I could not understand what you wrote, sorry ! do you mean I get z-socre for all activated pathway and simply make a bar plot for them ? or what
The list of genes you used came from some experiment correct? Then it must have original data associated with the gene names (e.g. counts, normalized values, log ratios etc). IPA allows you to import those values in. You will need to re-do the analysis though (if you had not imported those values in initially). This would be purely cosmetic (to color the gene names in pathways).
@genomax2 how can you find a pathway without value ??? for sure you MUST import them and I did import them. The genes are not important for me, I found few pathways that are activated and that is the main reason of my study. I want to know how to show that they are activated . based on IPA they are coloured by yellow (those that have z-score) , What I want is only to show that few pathways are activated , is this clear?
You don't need to have values for mapping a list of genes onto pathways. But I see what you are saying.
If you are referring to "Canonical pathways" or "Diseases and Functions" graphics/tables both of those are exportable. You can get a table for "Diseases and functions" with the p-values and activation z-scores and also an image.
@genomax2 it is true, however, if you have so many pathways , then it is not good for publications. I am talking about canonical pathways which shows those that have z-score positive activated and those that have z-score negative not activated. The problem is showing the results . lets say if you have 400 pathways , you cannot just pick up 10 and say these are activated no? also if I mention z-score, do you think is a good way to say it is activated ?
Don't have a good answer. Are a subset of results supporting specific finding(s) you are concentrating on in the paper? Perhaps you could include just those in paper body and say that there were others (and include the full list in supplementary materials).
@genomax2 yes , that is exactly what i wanted to do because apparently there is not any better way to say it. So probably the best is to put all identified pathways , z-score, p value etc in supplementary material and just mention that those pathways we wanted are also activated. One question , go to Canonical pathway , then click on one of the activated pathway you have , then below you can see the genes which are assigned to that pathway (up and down regulated) then in the right corner you can click on "Open Pathway" which shows the pathway. I saw in one tutorial that it is based on the analysed data . my mean if an enzyme showed in this pathway to be down regulated it is true for the data set analysed. Is this right? did i understand it correctly ?
genomice - intentional?
@Ram what is genomice international ? I use IPA because I think it is the most robust pathway analysis tool
"intentional", not "international". Your tags have a typo, and I asked if it was intentional as you may have been working on mouse genomics :)
@Ram thanks yep was a typo but actually it is relevant for mouse genomics too :-)