Hi all,
I have what I thought was a relatively simple question, but I haven't been able to find a simple solution..
We have a gene which is relatively unknown and we would like to understand in which conditions/diseases it is differentially regulated using unbiased mining of GEO. We have done this manually for a number of datasets and come up with some very novel suggestions for processes in which our gene may be involved, but are keen to take a more systematic and efficient approach (with our limited R experience).
We were initially planning on downloading the GEO profile for gene x, extracting the values, running a t-test, then
Is there an existing package that does this, or is there a simple way to approach this (our attempts so far have been unsuccessful!)?
LM.
3 answers
Have you looked at Gene Expression Atlas maintained by EBI for your gene of interest ?
Atlas provide pre-computed information including Organism part where the gene is differentially under/over expressed, Cell line, Cell type, Disease state, Compound treatment etc.
For example:
Here is an example page for one of my favorite human gene RGS6.
Specific to your question disease states associated with this gene is available here.
Also adding a figure to depict published studies where the gene is over (red)/under (blue) expressed compared to the gene's overall mean expression level in the study.

Bioconductor has a GEOquery package that may be of use to you. http://bioconductor.org/packages/release/bioc/html/GEOquery.html
This is another interesting approach:
Log in to answer this question.