Convert affymetrix CHO-K1 chip labels into gene symbols/IDs

Question

[Rockbar]

I have downloaded some microarray data from https://www.ebi.ac.uk/arrayexpress/experiments/E-GEOD-30321/ which are microarray data from some CHO-K1 culturing. The microarray is signed: "Affymetrix Chinese Hamster Ovary (CHO)-specific custom array"

To be able to use the data within the samples I have to convert the affymetrix IDs into gene IDs or symbols. How can I do that? I searched in Ensembl Biomart, but "CHO" or "Cricetulus griseus" is not there. Can it be alternatively done with R and when with which annotation package?

Thank you very much.

Answer 1

[mastal511]

There is a link from the E-GEOD30321 page to the array design, which gives GenBank IDs for some of the probeIDs, although not all.

https://www.ebi.ac.uk/arrayexpress/files/A-GEOD-13791/A-GEOD-13791.adf.txt

[Rockbar]

Thanks. Unfortunately, there are only few IDs mapping to Genbank, as you said.

Is there a tool/list to convert Genbank IDs into Gene IDs?

Answer 2

[ivivek_ngs]

Take a look at this link? You need to take a deeper look into google. It will be good practise for you. Any id should be convertible pre or post processing.

[Rockbar]

Thanks for your tipp with Google. ;-) I will work through your link step by step, that could be a really good help.

DAVID I tried already, it does not provide Cricetulus griseus (CHO) and reports a mapping success, genbank to Entrez Gene ID, of under 20 % with other Cricetulus organisms. Biomart itself reports to be down at the moment. Ensemble Biomart does not provide Cricetulus at all.

The publication I mentioned above reports also an comment file with genbank_to_GeneIDs, but weirdly enough only for Affymetrix spot labels, which could not be mapped to genbank ids.

I have started with Bioconductor packages (annotationDbi, affy, etc.), but to get handy with that will take some time. Could not detect annotation packages for cricetulus so far.

[ivivek_ngs]

ah ok, yes then it is of a concern. Yes Bioconductor is a bit tricky but yes definitely you need to get more familiar with it. It takes some time but obviously not impossible. Take a look at the links here and here enter link description hereif you are working with annotationDbi

If you say about mapping all ids to gene names that might not be true. It might be possible that you will have gene names for all the probes unless the genome is properly characterized and the functional annotation is not done.

Also take a look at this paper which should include how the pathway analysis was done or the gene set enrichment

[Rockbar]

Thanks. That is great! :-) I have also found "AnnotationHub" recently. Let us see how far I can get with that. Obviously, there is no Swiss army knife for my challenge.

[ivivek_ngs]

If the answers are of help you can accept them with upvotes and validate them as answers.