Question

Loading beta values into minfi

0

Entering edit mode

CC ▴ 10

@cc-8931

Last seen 5.8 years ago

United States

Hi there,

I have only beta values available to me in a matrix from 450k array data. I am trying to load the values into minfi as a GenomicRatioSet but keep getting the same error:

> ratioSet <- ratioConvert(betaM, what="beta", keepCN=TRUE)

> Error in (function (classes, fdef, mtable) : unable to find an inherited method for function ‘ratioConvert’ for signature ‘"matrix"’

I tried to make the matrix a dataframe instead and I get the same error except with the word "matrix" replaced with "data.frame".

I would like to use the getSex check in minfi and this requires loading in the GenomicRatioSet.

Any help you can provide would be greatly appreciated.

minfi 450k • 3.0k views

ADD COMMENT • link updated 8.1 years ago by Kasper Daniel Hansen 6.5k • written 8.1 years ago by CC ▴ 10

score 1 · Answer 1 · 2015-10-29

1

Entering edit mode

Kasper Daniel Hansen 6.5k

@kasper-daniel-hansen-2979

Last seen 5 months ago

United States

ratioConvert is not the right function, that one converts a MethylSet into a RatioSet, You want to directly create a RatioSet. Do this by tmp = RatioSet(Beta = betaM) (you can also supply CN estimates if you have those.). Then you have to set the annotation correctly, the easiest is to library(minfiData) annotation(tmp) = annotation(MsetEx) Then you convert to GenomicRatioSet by tmp2 = mapToGenome(tmp) Best, Kasper On Thu, Oct 29, 2015 at 3:13 PM, Sally [bioc] <noreply@bioconductor.org> wrote: > Activity on a post you are following on support.bioconductor.org > > User Sally <https: support.bioconductor.org="" u="" 8931=""/> wrote Question: > Loading beta values into minfi <https: support.bioconductor.org="" p="" 73941=""/>: > > > Hi there, > > I have only beta values available to me in a matrix from 450k array data. > I am trying to load the values into minfi as a GenomicRatioSet but keep > getting the same error: > > > ratioSet <- ratioConvert(betaM, what="beta", keepCN=TRUE) > > > Error in (function (classes, fdef, mtable) : unable to find an > inherited method for function ‘ratioConvert’ for signature ‘"matrix"’ > > I tried to make the matrix a dataframe instead and I get the same error > except with the word "matrix" replaced with "data.frame". > > I would like to use the getSex check in minfi and this requires loading in > the GenomicRatioSet. > > Any help you can provide would be greatly appreciated. > > > > ------------------------------ > > Post tags: minfi, 450k > > You may reply via email or visit Loading beta values into minfi >

ADD COMMENT • link 8.1 years ago Kasper Daniel Hansen 6.5k

0

Entering edit mode

Thank you so much for your fast and very helpful reply.

ADD REPLY • link 8.1 years ago CC ▴ 10

0

Entering edit mode

Hi Kasper,

I was able to load the data into a GenomicRatioSet using your advice. However, when I run GetSex, I get the following error - presumably because I do not have CN values available to me. Is there a way around this?

> Error in .getSex(CN = CN, xIndex = xIndex, yIndex = yIndex, cutoff = cutoff) :

must provide CN, xIndex, and yIndex

ADD REPLY • link 8.1 years ago CC ▴ 10

0

Entering edit mode

Well, do you have the methylation and unmethylation values or do you only have the beta values? On Tue, Nov 3, 2015 at 12:00 PM, Sally [bioc] <noreply@bioconductor.org> wrote: > Activity on a post you are following on support.bioconductor.org > > User Sally <https: support.bioconductor.org="" u="" 8931=""/> wrote Comment: > Loading beta values into minfi > <https: support.bioconductor.org="" p="" 73941="" #74078="">: > > Hi Kasper, > > I was able to load the data into a GenomicRatioSet using your advice. > However, when I run GetSex, I get the following error - presumably because > I do not have CN values available to me. Is there a way around this? > > > Error in .getSex(CN = CN, xIndex = xIndex, yIndex = yIndex, cutoff = > cutoff) : > > must provide CN, xIndex, and yIndex > > ------------------------------ > > Post tags: minfi, 450k > > You may reply via email or visit > C: Loading beta values into minfi >

ADD REPLY • link 8.1 years ago Kasper Daniel Hansen 6.5k

0

Entering edit mode

Just beta values unfortunately.

ADD REPLY • link 8.1 years ago CC ▴ 10

0

Entering edit mode

It is not impossible that you can do sex estimation by considering the beta values on the X-chromosome, since females (in general) ought to have higher methylation because one copy is silenced. But that is not what we have implemented and tested in minfi; to use the existing tools you need the methylation/unmethylation values. Best, Kasper On Tue, Nov 3, 2015 at 12:29 PM, Sally [bioc] <noreply@bioconductor.org> wrote: > Activity on a post you are following on support.bioconductor.org > > User Sally <https: support.bioconductor.org="" u="" 8931=""/> wrote Comment: > Loading beta values into minfi > <https: support.bioconductor.org="" p="" 73941="" #74083="">: > > Just beta values unfortunately. > > ------------------------------ > > Post tags: minfi, 450k > > You may reply via email or visit > C: Loading beta values into minfi >