[BioC] edgeR dispersions

[Mark Robinson]

Hi Bogdan,

> thanks for your email : it was possible to use edgeR in the past  (previous
> versions)  with no replicates,
> (so of course there is no dispersion), but edgeR worked ..while now I am
> getting errors ... thanks,

The default has changed to make users make a conscious decision of what to do about having no replicates.  The previous default (setting dispersion to a low value) was probably not ideal.

As Alessandro says, read the "What to do if you have no replicates" section of the user's guide:
http://www.bioconductor.org/packages/release/bioc/vignettes/edgeR/inst/doc/edgeRUsersGuide.pdf

Best,
Mark


----------
Prof. Dr. Mark Robinson
Bioinformatics
Institute of Molecular Life Sciences
University of Zurich
Winterthurerstrasse 190
8057 Zurich
Switzerland

v: +41 44 635 4848
f: +41 44 635 6898
e: mark.robinson at imls.uzh.ch
o: Y11-J-16
w: http://tiny.cc/mrobin

----------
http://www.fgcz.ch/Bioconductor2012

On 07.07.2012, at 09:07, Bogdan Tanasa wrote:

> Hi Alessandro,
> 
> thanks for your email : it was possible to use edgeR in the past  (previous
> versions)  with no replicates,
> (so of course there is no dispersion), but edgeR worked ..while now I am
> getting errors ... thanks,
> 
> Bogdan
> 
> 
> 
> 
> On Sat, Jul 7, 2012 at 12:02 AM, Guffanti Alessandro <
> alessandro.guffanti at genomnia.com> wrote:
> 
>> Hi Bogdan - you answered by yourself: you don't have replicates, so how
>> can any software estimate any kind of (common or tagwise) dispersion ?
>> there is one section on edgeR manual with interesting hints about what you
>> can - and can not - evaluate with such 'minimal' experiments
>> 
>> My two cents, regards,
>> 
>> Alessandro
>> -----------------------------------------------------
>> *Alessandro Guffanti - Bioinformatics, Genomnia srl*
>> Via Nerviano, 31 - 20020 Lainate, Milano, Italy
>> Ph: +39-0293305.702 Fax: +39-0293305.777
>> http://www.genomnia.com
>> "If you can dream it, you can do it" (Walt Disney)
>> 
>> -----Original Message-----
>> From: Bogdan Tanasa <tanasa at gmail.com>
>> To: <bioconductor at stat.math.ethz.ch>
>> Date: Fri, 6 Jul 2012 15:57:13 -0700
>> Subject: [BioC] Fwd: edgeR dispersions
>> 
>> Dear all,
>> 
>> I would appreciate a suggestion about edgeR : I have a GRO-seq dataset (CTL
>> vs TREAT, no replicates, just 1 sample for CTL and 1 sample for TREAT
>> condition),
>> and I am assessing the differential expression by using edgeR, but I am
>> getting an error : any suggestion would be appreciated. thanks very much !
>> 
>>> d <- estimateCommonDisp(d)
>> Warning message:
>> In estimateCommonDisp(d) :
>>  There is no replication, setting dispersion to NA.
>> 
>>> de.com <- exactTest(d)
>> Error in if (any(dispersion < 0)) stop("dispersion must be non-negative") :
>>  missing value where TRUE/FALSE needed
>> 
>>> de.com <- exactTest(d,dispersion=0)
>> Error in binomTest(s1[pois], s2[pois], p = n1/(n1 + n2)) :
>>  y1 and y2 must be non-negative
>> 
>>   [[alternative HTML version deleted]]
>> 
>> _______________________________________________
>> 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
>> 
>> 
>> -----------------------------------------------------------
>> Il Contenuto del presente messaggio potrebbe contenere informazioni
>> confidenziali a favore dei
>> soli destinatari del messaggio stesso. Qualora riceviate per errore questo
>> messaggio siete pregati
>> di cancellarlo dalla memoria del computer e di contattare i numeri sopra
>> indicati. Ogni utilizzo o
>> ritrasmissione dei contenuti del messaggio da parte di soggetti diversi
>> dai destinatari è da
>> considerarsi vietato ed abusivo.
>> 
>> The information transmitted is intended only for the p...{{dropped:15}}
> 
> _______________________________________________
> 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