Illumina Microarray Annotation

Question

[andrew.j.skelton73]

When Analysing Illumina (HT-12) microarrays, I always tend to find that there are NAs returned in some cases (gene name and description) as being differentially expressed. How is something determined to be an NA, I thought they Illumina database would have an entry for every probe sequence used.... Any explanations?

Thanks

[Istvan Albert]

what software are you using?

[andrew.j.skelton73]

Bioconductor -> get nuIDs -> query for gene name and description

Answer 1

[Ying W]

NAs are typically unannotated genes or bad probes. Are you using Genome Studio or HuRef annotations on R?

[andrew.j.skelton73]

Bioconductor -> get nuIDs from probe -> query for gene name and description

[andrew.j.skelton73]

I think you're right. If there's 47K probes on the Illumina HT-12 arrays, There are ~20K protein coding genes in the human genome, 13K Pseudo Genes, I guess that anything that's not associated with a gene is not annotated. I pulled out a few NA entries and their probe sequences, and they all mapped to Pseudogenes on the human genome

[Ying W]

Just a sidenote but keep in mind that you are relying on whoever annotates nuIDs to go from probes->genes. Different annotation sources are in different states of outdated-ness. I know some of the genes on HT-12 used to be annotated as pseudo-genes but then were found not to be real so you could sometimes get the case of depreciated Entrez IDs (depending once again on what annotation source you use).

Answer 2

[Ashutosh Pandey]

You can blat the probe sequence against the genome in UCSC genome browser and that can help in some cases.