What to do with the empty rows of results obtained with the Affymetrix Expression Console?

Question

[Hamtaro]

Hello everybody.

This is the first time I perform an HTA 2.0 microaray analysis using the Affynmetry Console.

Once the analysis is finished, I get empty rows in the Summary View and Gene View table in Gene Symbol column. Is there any way to identify that probe or do you filter out those that have not been identified?

This is an example:

Thank you in advance

[Kevin Blighe]

Can you confirm the input files that you used? For example, which NetAffx file?

[Hamtaro]

I used GeneChip™ Human Transcriptome Array 2.0 microarrays. The raw files are .CEL files

Answer 1

[Kevin Blighe]

Some, or perhaps all, of these seem to relate to protein coding genes. The reason why they have no annotation could be because they were not yet assigned official gene symbols when this array was manufactured. Some of these already have Ensembl IDs, though.

There is extra information about them in the Net Affx file for this array: https://www.thermofisher.com/order/catalog/product/902162#/902162

grep -e 'TC07000339' HTA-2_0.na36.hg19.probeset.csv 
"PSR07005056.hg.1","chr7","+","56123058","56123195","30","TC07000339.hg","Coding","EX07027581.hg","PSR07005056.hg.1","---","ENST00000383876 // chr7 // 100 // 0 // --- // 1 /// uc022adt.1 // chr7 // 100 // 0 // --- // 1","main","1","TR07002358.hg","ENS//UCG","Constituitive","1","100.00 %","pass","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---"

grep -e 'TC15002026' HTA-2_0.na36.hg19.probeset.csv 
"PSR15018739.hg.1","chr15","-","102294519","102294548","6","TC15002026.hg","Coding","EX15013431.hg","PSR15018739.hg.1","---","DQ575740 // chr15 // 100 // 0 // --- // 1 /// DQ575740 // chr15 // 100 // 0 // --- // 1 /// uc021syd.1 // chr15 // 100 // 0 // --- // 1","main","1","TR15007323.hg","UCG","Constituitive","1","100.00 %","pass","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---","---"

I would download the Net Afffx file and find some way to add the annotation. As you are using the Affymetrix Expression console (?), there must be a way to add extra fields from the Net Affx file. Otherwise, you could simply perform the analysis on the command line, which would be easier in many respects.

Kevin

[Hamtaro]

I just check what you said. I'm using the last annotation file updated in TAC catalogue (https://freeimage.host/i/Ju25Rn)

This is very weird, because analyzing the same CEL with Bioconductor, in this case it does find the name of the gene.

[Kevin Blighe]

By the way, I show, through the following link, how one can create their own annotation database package from a Net Affx file: https://support.bioconductor.org/p/130727/#130733