Entering edit mode
5.7 years ago
indolefin
•
0
Hi all
I am looking for a tool/web-service to compare two sets of cell lines, e. g. one that is resistant and the other being sensitive. I want to know whether there are mutations or alterations in expression levels that are common or at least more frequent in either the first or the second group of cell lines. So far I found http://cellexpress.cgm.ntu.edu.tw/ where one can compare groups. The heatmaps I got by the "Gene Signature Explore" function look promising. However, I can't find which genes are significantly differently expressed between the two groups. Do you have any ideas which tool or web-service can provide the differences of two sets of cell lines?
What data do you have? Finding mutations and gene expression etc are different tasks that different types of data are optimal for.
Thanks for your question! I only know the names of the cell lines but most of them are included in some databases. From the databases I could retrieve further data like mutations and expressions. So in principle I could compare them manually but this would be very time consuming and more error prone than having a tool doing the task.
Its still not clear to me what you want to do?
You want to magic up a list of all the ways in which 2 cell lines are different? That's not going to happen as far as I know.
You need a more defined question; i.e. "Is cancer cell line A's transcription under condition X different in some specific way from Cancer line B's".
I want to know whether there is an explanation why two sets of cell lines behave differently when treated with the same substance. This explanation might be for example an over expression of a transporter protein or a mutation of a metabolizing enzyme. If it is too hard to compare expression levels as they massively depend on the environmental conditions of the cells (and I do not have the expression profiles when treated with this substance) it would be nice to know mutations that are more common along one of the sets, e. g. mutations that drive resistance.
Referring to your example: The question would be: Which genes of cell lines of group A are under normal conditions at least x-fold under or over expressed with p -value < y when compared to the transcription of group B cell lines?
The ideal tool would receive the names of the cell lines of each group as input and output something like gene A is mutated or over expressed in 80% of group A but only in 5% of group B.
Is it now clearer what I want to do? I am just starting working with cellular data so maybe I am just missing something very basic.
I very much doubt any such tool exists. I think quite the contrary, what is being missed is far from basic.
The best way I can think to do this is just by mining the literature to look for studies which have shown responses to this compound. Others might know more though.
You couldn’t even reliably reanalyse the data for yourself if different transcripts mics data came from different experiments as the variance would be enormous.
Thank you very much for your reply!
I think I do not necessarily need the responses to the compound because the response is probably at least partly predefined by the transcription profile before the treatment with the compound. The compound is only exposed to the cells for a few days. This time is presumably to short to develop a resistance. So, the cells are maybe resistant due to the treatment of the patient they originate from with a drug.
This is something you will need to experimentally determine yourself I think. I doubt you will find the answers neatly packaged online..
Thanks for your help!
Hi indolefin,
I am currently working on the same kind of question/hypothesis that you had. Did you find a tool to compare your groups of cell lines?