[BioC] using a gtf file to map reads

Michael Love michaelisaiahlove at gmail.com
Tue Jun 11 18:39:32 CEST 2013


hi Valerie,

On Tue, Jun 11, 2013 at 6:13 PM, Valerie Obenchain <vobencha at fhcrc.org> wrote:
> Hi Mike (Love),
>
> Would you be interested in contributing a disjointExons() to
> GenomicFeatures? I think this extraction would be useful to many. Maybe we
> also want a disjointExonsBy() but the only 'by' would be genes ...?
>
> Valerie
>

I'd be happy to contribute any of this code in parathyroidSE.
Alejandro Reyes has formalized this part of the code into the
following function in DEXSeq >= 1.6.0:

> prepareAnnotationForDEXSeq
function (transcriptDb, aggregateGenes = FALSE, includeTranscripts = TRUE)
{
    stopifnot(is(transcriptDb, "TranscriptDb"))
    exonsByGene <- exonsBy(transcriptDb, by = "gene")
    exonicParts <- disjoin(unlist(exonsByGene))
    if (!aggregateGenes) {
        overlaps <- findOverlaps(exonicParts, exonsByGene)
        geneNames <- names(exonsByGene)[subjectHits(overlaps)]
.....

best,

Mike



More information about the Bioconductor mailing list