To compare gene expression profiles.

Question

[friasoler]

Hello

Thanks to all who help me answering my questions!

I have huge list of genes from different tissues and conditions and I would like to compare these profiles with some other results. I'm trying to find which genes/metabolic pathways from my list could be "new developed" in each tissue. Usually all approaches are designed to find from a list of genes de ontology...etc but how can I relice among 11000 genes which one has not found before in that tissue?

My best regards and thanks in advance

Roberto

[Sean Davis]

What do you mean by "which ones have not been found before in that tissue"? Where do the metabolic pathways come in? A little more context would be useful.

[elia.brodsky]

Hi Roberto, is this similar to this discussion: Retrieving Gene Expression Profiles For A Set Of Genes From All Tissues?

[friasoler]

oh thanks ...but no really .-(

Answer 1

[elia.brodsky]

Do you mean "differential expression" or identifying which genes are expressed differently in different conditions or in different tissues? If so, here is an article that explains some of the common algorithms that are used: http://www.genomebiology.com/content/pdf/gb-2010-11-12-220.pdf

[friasoler]

Hello

Thanks for answer me...I have already done the "differential expresion" in my data; but I would like to compare the genes I have found in each tissue (qualitativelly) with another genetic profiles. I want to compare the genes expressed in liver from my experiments with another reports of gene expression in liver. As I have RNASeq data im playing with 10000 genes per tissue.

Thanks in advance for helping me

Roberto

[Sam]

So if we understand correctly, what you want to do is to identify the novel changes in your experiment that has never been found before, right?

Then my question to you will be: Do you want to identify gene that was not previously found to be expressed in liver? Or do you want to identify DE genes that was never reported in your condition/in liver?

[Sean Davis]

Have a look at MSigDb.

[friasoler]

Dear Sam thanks for your comment

Yes both things will be great; but if at least the qualitative information is avaible I will be very happy! If I could discard thousands of genes that are commonly expressed in liver in vertebrates willl be more easy to find out exiting genes in my own species.

kind regards
Roberto

[Sam]

So basically you want to identify the gene expression specific to your species?

But then gene expression isn't a static thing, so it is likely that at some point the gene is expressed and in some other occasion, it is not.

[friasoler]

Yes sure but I want to discard the hausekeeping genes per tissue...and focous in the ones that...change .-) remember my list have 11000 genes!!!!

roberto