[BioC] Testing for differential expression variability with edgeR

[Gordon K Smyth]

Dear Miguel,

In principle, constructing a likelihood ratio test of equality of 
dispersions for two groups would not be difficult.  I'd have to sit down 
and work out some code, however, and I don't have time to do that just 
now.  It is unclear to me whether it could be done reliably for just two 
libraries in each group.  This is first time I've been asked for it.

Best wishes
Gordon

Professor Gordon K Smyth,
Bioinformatics Division,
Walter and Eliza Hall Institute of Medical Research,
1G Royal Parade, Parkville, Vic 3052, Australia.
Tel: (03) 9345 2326, Fax (03) 9347 0852,
smyth at wehi.edu.au
http://www.wehi.edu.au
http://www.statsci.org/smyth

> On Thu, Feb 23, 2012 at 1:19 AM, Gordon K Smyth <smyth at wehi.edu.au> wrote:
>
>> Dear Miguel,
>>
>> Well, you could obviously subset your data, and compute the tagwise
>> dispersions separately for each subset, and compare the two.  Do a formal
>> test requires a more work, and would be genewise anyway.
>>
>> I find it hard to imagine in what context it would make sense to test
>> whether two libraries (R4.Hot and R5.Hot) are more different than two other
>> libraries.  For a particular gene?  Averaged over genes?
>>
>> Could you not just make an MDS plot and look to see how far apart the
>> samples are from each other?
>>
>> What are you trying to achieve here from a biological point of view?  What
>> biological question are you trying to answer?
>>
>> Best wishes
>> Gordon
>>
>>  Date: Tue, 21 Feb 2012 15:16:53 +0100
>>> From: Miguel Gallach <miguel.gallach at univie.ac.at>
>>> To: Bioconductor mailing list <bioconductor at r-project.org>
>>> Subject: [BioC] Testing for differential expression variability with
>>>        edgeR
>>>
>>> Dear Bioconductor list,
>>>
>>> I am using edgeR to test for differential expression but I also I would
>>> like to know how to test whether expression variability (per gene) is
>>> significantly different between biological groups or treatments.
>>>
>>> For instance, the next is my experimental design, according to which I
>>> have
>>> two treatments (Hot and Cold) and two biological groups (Hot Adapted and
>>> Cold adapted; two replicates each).
>>>
>>> $samples
>>>                group lib.size norm.factors
>>> R4.Hot     HotAdaptedHot 17409289    0.9881635
>>> R5.Hot     HotAdaptedHot 17642552    1.0818144
>>> R9.Hot    ColdAdaptedHot 20010974    0.8621807
>>> R10.Hot   ColdAdaptedHot 14064143    0.8932791
>>> R4.Cold   HotAdaptedCold 11968317    1.0061084
>>> R5.Cold   HotAdaptedCold 11072832    1.0523857
>>> R9.Cold  ColdAdaptedCold 22386103    1.0520949
>>> R10.Cold ColdAdaptedCold 17408532    1.0903311
>>>
>>>
>>> I would like to detect those genes for which the expression variability
>>> between R4.Hot and R5.Hot is significantly different to that obtained for
>>> R4.Cold and R5.Cold. Can I use the tagwise dispersion values provided by
>>> edgeR, and how??
>>>
>>> Many thanks!
>>> Miguel
> -- 
> Miguel Gallach
> Center for Integrative Bioinformatics Vienna (CIBIV)
> Max F. Perutz Laboratories(MFPL)
> Telf: +43 1 4277 24029
>
> Postal Address:
> Ebene 1
> Campus Vienna Biocenter 5
> CIBIV, MFPL
> 1030 Vienna
> Austria
>
> e-mail:
> miguel.gallach at univie.ac.at
> migaca2001 at gmail.com
>

______________________________________________________________________
The information in this email is confidential and intend...{{dropped:4}}