Michael, thank you very much for the explanation! I will definitely try the weighted coverage option. Thanks again and all the best! Ales Maver, MD Institute of Medical Genetics, Department of Obstetrics and Gynaecology UMC Ljubljana Šlajmerjeva 3 SI-1000 Ljubljana Slovenia Tel/ fax +38615401137 GSM: +38631710721 E-mail: Ales.Maver@gmail.com <ales.maver@gmail.com> 2012/6/27 Michael Lawrence <lawrence.michael@gene.com> > > > On Wed, Jun 27, 2012 at 11:39 AM, Aleš Maver <ales.maver@gmail.com> wrote: > >> Dear BioConductors! >> >> I have been using functions in GenomicRanges package to find overlaps >> between multiple genomic tracks stored as GRanges objects. In the metadata >> section of the GRanges object, I store the feature's score data and I have >> been wondering if you can perform any operations with this data. For >> example, I would like to perform a sliding window analysis, where score >> values would be averaged for every window of a defined range? Is this >> possible using GenomicRanges? >> Even if I use the reduce function, the metadata gets lost from the >> resulting object - is there any possibility to retain or operate on >> metadata? >> >> > The metadata is discarded, since the features may have been fundamentally > altered by the reduce operation. > > As far as window-based computations, there is no easy way. If your data > makes sense as a genome-length vector, with zeros between the features, you > could use coverage(gr, weight = "score") to get the score as an RleList. > Then use functions like runmean(). If zeros in the gaps do not make sense, > you will need to somehow set them to NA to exclude them. > > Michael > > >> Thank you for any input on this! >> >> I wish you all a good day, >> Ales >> >> ****************** >> Ales Maver >> Institute of Medical Genetics, Department of Obstetrics and Gynaecology >> UMC Ljubljana >> Šlajmerjeva 3 >> SI-1000 Ljubljana >> Slovenia >> Tel/ fax +38615401137 >> GSM: +38631710721 >> E-mail: Ales.Maver@gmail.com <ales.maver@gmail.com> >> >> [[alternative HTML version deleted]] >> >> >> _______________________________________________ >> 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 >> > > [[alternative HTML version deleted]]