Extracting a list of genes from the Gene Expression Omnibus with Geoquery.
1
0
Entering edit mode
Caitlin ▴ 50
@caitlin-5400
Last seen 18 days ago
United States

Hi all.

I initially posted this on StackOverflow, but I did not receive any responses other than a "downvote". I am hoping I fare better here.

Is there a convenient way to extract a list of genes from a microarray experiment on the Gene Expression Omnibus? The following paper is serving as a source of inspiration and I would like to use it as a model:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5661398/

I have already located one GSE entry and I am searching for another that explores chemotherapy resistance. My intention is a perform a gene ontology using a tool like DAVID, identify what genes are differentially expressed, and constructing a graph with Cytoscape or RCy3. If there is another library that would facilitate my objective, I would be quite eager to learn about it.

Thank you.

Caitlin

Code should be placed in three backticks as shown below


# include your problematic code here with any corresponding output
# please also include the results of running the following in an R session

sessionInfo( )

Microarray GEOquery cancerdata MicroarrayData • 199 views
1
Entering edit mode
@james-w-macdonald-5106
Last seen 22 hours ago
United States

Have you read the GEOquery vignette? That should get you 99% of the way there. Note that the fData slot of an ExpressionSet contains all the gene information, so you can get a list of all the genes on the array using that.

I should point out that you appear to have things backwards though. To do a GO test, you first get a list of differentially expressed genes (using for example the limma package), and then you do the GO testing. There is really no reason to use DAVID for that. There are the GOstats and topGO packages, as well as the goana function within limma, all of which do various forms of GO hypergeometric testing.