Ok I see. DESeq2 doesn't provide this functionality, sorry! The design matrix is fixed for all rows/genes. On Mon, Jun 30, 2014 at 11:59 AM, Jorge Zamora <jz1@sanger.ac.uk> wrote: > Hello Michael, > > For each sample and region I have the percentage homozigosity. I would > ideally have the formula specified as counts ~ treatment + homozigosity. > > Basically I have two variables that are specified per sample and region > which are the 'counts' (dependent variable) and the 'homozigosity > percentage' (independent variable), and then I have the 'treatment' which > depends only on the sample. > > I understand I could potentially perform a NB GLM on such dataset but I > would loose all the features provided by DESeq2. > > Any advise will be much appreciated. > > Thanks > Jorge > > > > > On 30 Jun 2014, at 16:25, Michael Love <michaelisaiahlove@gmail.com> > wrote: > > hi Jorge, > > DESeq2 does not provide functionality to perform modeling with a different > design matrix for each row (genomic range). > > Can you tell us more about the experiment you are trying to analyze? > > Section 1.5 is discussing design with multiple factors, e.g.: counts ~ > group + treatment, this is not related to your question I think. > > Mike > > > On Mon, Jun 30, 2014 at 11:18 AM, Jorge Zamora <jz1@sanger.ac.uk> wrote: > >> Hello, >> >> I'm using DESeq2 and I would be interested in specifying continuous >> independent variables for each sample and genomic range. I see in section >> 1.5 (page 10) of the documentation that is possible to associate a given >> sample with a numeric value. However such values are fixed for all the >> genomic regions in that sample. >> >> Any help will be highly appreciated. >> >> Thanks a lot >> Jorge >> >> -- >> The Wellcome Trust Sanger Institute is operated by Genome Research >> Limited, a charity registered in England with number 1021457 and a >> company registered in England with number 2742969, whose registered >> office is 215 Euston Road, London, NW1 2BE. >> >> _______________________________________________ >> Bioconductor mailing list >> Bioconductor@r-project.org >> https://stat.ethz.ch/mailman/listinfo/bioconductor >> Search the archives: >> http://news.gmane.org/gmane.science.biology.informatics.conductor >> > > > > -- The Wellcome Trust Sanger Institute is operated by Genome Research > Limited, a charity registered in England with number 1021457 and a company > registered in England with number 2742969, whose regis...{{dropped:7}}