[HuEx-1_0-st] Affymetrix Human Exon 1.0 ST Array [transcript (gene) version]

Question

[vinnu260]

Hi,

I'm working on Exon arrays i.e.. [HuEx-1_0-st] Affymetrix Human Exon 1.0 ST Array [transcript (gene) version] - GPL5175. I did RMA with target="core" option using oligo package. i got 22011 probe clusters, which I'm trying to annotate.

Is the procedure correct for normalization? Which package should i use to annotate this data?

Thanks in advance

Answer 1

[Kevin Blighe]

Yes, well, target="core" typically summarises the expression values into gene-level values; whereas, target="probeset" will leave the expression values at the level of probe-sets. One or many probe-sets may target an individual exon, for example.

As you're using an 'Exon' array, however, I believe target="core" will still leave your values at the level of exons. For these arrays, two other options are available:

• target = ’full'
• target = ’extended’

----------------------------------

Edit (21st July 2018):

For annotation, see the working example here: A: Affymetrix Human Genome U133 Plus 2.0 Array

Also see the biomaRt vignette (section 4.1).

-----------------------------------

Kevin

[vinnu260]

Thanks for the response.

Can you please tell me the difference between 1) [HuEx-1_0-st] Affymetrix Human Exon 1.0 ST Array transcript (gene) version vs [HuEx-1_0-st] Affymetrix Human Exon 1.0 ST Array probe set (exon) version ? Im using the transcript version though both are available. 2) HuGene-1_0-st-v1 Transcript Cluster Annotations vs HuGene-1_0-st-v2 Transcript Cluster Annotations ? I tried using the v2 version of the same and ended up with clusters mapping to mutiple genes including miRNAs and few thousands of control clusters.

After annotating with HuGene-1_0-st-v2 Transcript Cluster Annotations file and removing clusters mapping multiple genes/control probes i ended up with 15500 genes out of 22011 transcript clusters. I expected them to map one to one at transcript level. Does this make sense or am i doing it wrong ?

The thread you mentioned talks about [HuGene-1_1-st] Affymetrix Human Gene 1.1 ST Array transcript (gene) version. Is there any other way of annotating transcript version of this exon arrays(GPL5175 and GPL5188) ? This article(https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5263238/) uses GPL5175 and GPL5188 arrays but they got 26,493 transcriptswhich is way more than what im having(15500 transcripts). Can you help me understand how the array yielded 26000 transcripts ?

Thank you for the help.

[Kevin Blighe]

[vinnu260]

Sorry for the wrong name. It was HuEx-1_0-st-v2 Transcript Cluster Annotations, CSV, Release 36 (95 MB, 7/6/16) (http://www.affymetrix.com/support/technical/byproduct.affx?product=huexon-st) i.e.. GPL5175. Can you share the R code you are using to preprocess ? Thank You.

[Kevin Blighe]

Do you still need this?

[vinnu260]

Yeah if an answer is available

[Kevin Blighe]

[vinnu260]

Thank you. I did the same but I'm facing problem with annotation of probes. Kindly check previous comments.

[Kevin Blighe]

[Kevin Blighe]

[vinnu260]

Ill try to do the same with human 1st array. Thanks