Question

Error while running affylmGUI

0

Entering edit mode

devyani66 • 0

@devyani66-9451

Last seen 5.7 years ago

Hello, I am trying to run the affylmGUI R package and I get this error on CDF env during the normalization step. Below is the link: CDF env Error

Any idea what might be the reason? I googled it but didnt find much help. Thanks in advance.

microarray limma affy affylmGUI • 1.5k views

ADD COMMENT • link updated 5.7 years ago by Gordon Smyth 51k • written 5.7 years ago by devyani66 • 0

Gordon Smyth · Answer 1 · 2019-02-27

1

Entering edit mode

Gordon Smyth 51k

@gordon-smyth

Last seen 2 hours ago

WEHI, Melbourne, Australia

The error message tells you what the problem is. You need install the rmhu01aa520485cdf CDF package so that affylmGUI knows how your arrays are designed.

The problem is that you are trying to analyse data from a custom Rosetta-Affymetrix platform and Bioconductor doesn't have a standard CDF package for this array: https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GPL14189

A similar question was asked about 5 years ago and James MacDonald explained how to get the package: https://support.bioconductor.org/p/59723

This question and response from 8 years ago is also relevant: https://support.bioconductor.org/p/40455/

I wonder what you Googled? I searched for "rmhu01aa520485cdf" and James' answer came up straight away.

ADD COMMENT • link 5.7 years ago Gordon Smyth 51k

0

Entering edit mode

Thank you for your reply. My bad I didnt mention that I did go through the above link where MacDonald explained how to get the package. I had tried everything mentioned in and still, it used to give the same error.

Now going through again I realized the problem was with the name of the cdf file. If you see the discussion, in many places rmhu01 is mentioned as rmhu10. That simple change of 01 to 10 was the issue as even though my env was created but it was rmhu10aa520485cdf and not rmhu01aa520485cdf- the one its searching for. I had just copied the command from there and changed the file name (.gz) and kept the rest same...so this silly mistake. But to be honest that discussion thread is a big confusing one...with many things repeated and hard to understand which response is for which query.

But since you pointed again, I went back and realized this difference. Thank you.

ADD REPLY • link 5.7 years ago devyani66 • 0

0

Entering edit mode

Hi Mr.Gordon, I have one more query regarding this. So, I ran these steps before carrying out normalization using affylmGUI.

> library(makecdfenv)
> make.cdf.package("GPL10687_RM-HU01Aa520485_custom_MMPM.cdf.gz",
> "rmhu01aa520485cdf", compress=TRUE, species="Homo_sapiens")
> BiocManager::install("AnnotationDbi")
> install.packages("rmhu01aa520485cdf", repos=NULL, type="source")

I exported the normalized values but the file has just probe ids and no annotation. Is there something else too that I need to run to include annotation? Thanks in advance.

ADD REPLY • link updated 5.7 years ago by Gordon Smyth 51k • written 5.7 years ago by devyani66 • 0

0

Entering edit mode

I checked you answer to a similar question 8 yrs ago- https://support.bioconductor.org/p/40823/

I am not sure if I understood well, but I suppose I have to install the proper annotation package. For me, while I tried to install "rmhu01aa520485cdf" file, it wrote I did not have "AnnotationDbi" package and so I downloaded it.

ADD REPLY • link 5.7 years ago devyani66 • 0

0

Entering edit mode

You need to create CDF and annotation packages for your custom arrays. I have no experience with that. I can only answer questions about affylmGUI.

I suggest you ask a new question with an informative title and tags so that the right people will see your question.

But beware that affyLMGUI was never intended to be used with custom arrays, so you might have to consider using the affy and limma packages directly instead of through the GUI.

ADD REPLY • link 5.7 years ago Gordon Smyth 51k

0

Entering edit mode

Thank you for your response. And so sorry for a super late reply. To keep it simple, I just mapped the probe ids from the normalized file to the annotation file using simple R command. It worked. Regards

ADD REPLY • link 5.7 years ago devyani66 • 0