[BioC] RNA degradation plot with oligo package GeneFeatureSet objects

heyi xiao xiaoheyiyh at yahoo.com
Sat Aug 17 00:28:00 CEST 2013


Hi Christian,
Great, I will check that out. Thanks for the info.
Heyi

--------------------------------------------
On Fri, 8/16/13, cstrato <cstrato at aon.at> wrote:

 Subject: Re: [BioC] RNA degradation plot with oligo package GeneFeatureSet objects

 Cc: bioconductor at r-project.org
 Date: Friday, August 16, 2013, 4:32 PM

 Dear Heyi,

 Yes, xps should work with the Ovine Gene 1.1 ST arrays,
 since it does 
 work with the Human, Mouse and Rat Gene 1.1 ST arrays. These
 arrays use 
 the PGF-files from Affymetrix and
 'xps/examples/script4schemes.R' shows 
 you how to create the 'scheme' files for xps.

 Best regards,
 Christian


 On 8/16/13 9:36 PM, heyi xiao wrote:
 > Hi Christian,
 > Thanks for the suggestion and sharing of degradation
 plot experience. Does xps work with the Ovine Gene 1.1 ST
 array? I don’t have a CDF package.
 > Heyi
 >
 > --------------------------------------------
 > On Fri, 8/16/13, cstrato <cstrato at aon.at>
 wrote:
 >
 >   Subject: Re: [BioC] RNA degradation
 plot with oligo package GeneFeatureSet objects
 >   To: "James W. MacDonald" <jmacdon at uw.edu>,

 >   Cc: bioconductor at r-project.org
 >   Date: Friday, August 16, 2013, 2:01
 PM
 >
 >   Dear Heyi,
 >
 >   Function 'plotAffyRNAdeg()' of package
 'xps' does allow you
 >   to plot RNA
 >   degradation plots for Whole Genome and
 Exon arrays, see
 >   Cahpter 5.4.1 of
 >   vignette 'xps.pdf'.
 >
 >   I have done this for a couple of
 HuGene array data, and the
 >   results look
 >   interestingly. Especially there is a
 difference when you
 >   compare RNA
 >   degradation plots from frozen tissues
 vs paraffin embedded
 >   tissues.
 >   However, I am not sure how to
 interpret the results.
 >
 >   Best regards,
 >   Christian
 >   _._._._._._._._._._._._._._._._._._
 >   C.h.r.i.s.t.i.a.n   S.t.r.a.t.o.w.a
 >   V.i.e.n.n.a
 >      A.u.s.t.r.i.a
 >   e.m.a.i.l:       
 cstrato at aon.at
 >   _._._._._._._._._._._._._._._._._._
 >
 >
 >
 >   On 8/16/13 6:52 PM, James W. MacDonald
 wrote:
 >   > Hi Heyi,
 >   >
 >   > On 8/14/2013 4:47 PM, heyi xiao
 wrote:
 >   >> Hi all,
 >   >> In affy package, I can use
 AffyRNAdeg and
 >   plotAffyRNAdeg to plot and
 >   >> check RNA degradation. Is
 there any way to do so in
 >   oligo package for
 >   >> GeneFeatureSet,which is
 equivalent to AffyBatch in
 >   affy package. I
 >   >> look at the GeneFeatureSet
 and AffyBatch, they
 >   quite similar. But not
 >   >> sure what can be done here. I
 can either modify
 >   AffyRNAdeg and
 >   >> plotAffyRNAdeg functions to
 fit them for
 >   GeneFeatureSet, or I can
 >   >> convert GeneFeatureSet to
 AffyBatch and use the
 >   affy package
 >   >> degradation functions. Any
 suggestions would be
 >   highly appreciated.
 >   >
 >   > While I suppose you could
 hypothetically do the
 >   conversion, I wonder if
 >   > it makes conceptual sense.
 >   >
 >   > The 3'-biased Affy arrays were
 all based off an
 >   oligo-dT primer that was
 >   > used to convert mRNA to cDNA, so
 the reverse
 >   transcription proceeded
 >   > from the 3' end of the mRNA,
 always. In this case you
 >   can wonder about
 >   > two things. First, how far did
 the RT step proceed? Did
 >   you in general
 >   > get good RT all the way to the
 most 5' of the probes in
 >   the probesets?
 >   >
 >   > Second, since we were using the
 polyA tail at the 3'
 >   end, by definition
 >   > the mRNA wasn't degraded from the
 3' end. However, it
 >   might have had
 >   > more or less extensive
 degradation from the 5' end, so
 >   the RT may have
 >   > gone to completion, but the
 degradation had proceeded
 >   past the most 5'
 >   > probes.
 >   >
 >   > So both things are confounded, as
 we cannot distinguish
 >   RT that didn't
 >   > proceed too far from highly
 degraded mRNA, but no
 >   matter. What we could
 >   > do is say how much signal we were
 getting from the more
 >   5' probes, and
 >   > decide if we wanted to do
 something about that (like
 >   only use the first
 >   > 8 probes or whatever).
 >   >
 >   > For the newer generation of Affy
 arrays, we use a
 >   random primer, so the
 >   > RT step proceeds from a random
 point in the transcript
 >   and proceeds
 >   > towards the 5' end (at least I
 think it is still
 >   directional). Since the
 >   > RT no longer starts from one end
 of the transcript, it
 >   is no longer
 >   > clear what differential amounts
 of probe signal would
 >   actually signify.
 >   >
 >   > In addition, with the newer
 generation of Affy arrays,
 >   we can collapse
 >   > the probes into different
 probesets, depending on what
 >   we are trying to
 >   > measure (e.g., you can try to
 measure expression at the
 >   exon level or
 >   > the transcript level).
 >   >
 >   > I think trying to do this would
 be more difficult than
 >   it would be
 >   > worth, especially given that I
 don't know what you
 >   would do if you were
 >   > to decide there had been
 degradation.
 >   >
 >   > Best,
 >   >
 >   > Jim
 >   >
 >   >
 >   >> Heyi
 >   >>
 >   >>
 _______________________________________________
 >   >> Bioconductor mailing list
 >   >> Bioconductor at r-project.org
 >   >> https://stat.ethz.ch/mailman/listinfo/bioconductor
 >   >> Search the archives:
 >   >> http://news.gmane.org/gmane.science.biology.informatics.conductor
 >   >
 >
 >



More information about the Bioconductor mailing list