On Tue, Dec 14, 2010 at 1:43 PM, R. T. Sweeney <[email protected]> wrote:
> Hi,
> I am new to bioconductor and working with unannotated viral genomes and
> using some gene prediction algorithms to find
> gene-like regions in the genomes.
>
> In ShortRead, I am interesting in using the "Filtering input" modality
> (below) much like chromosome filtering but for gene filtering (all reads to
> specific gene region).
> Below, there is cfilt <- chromosomeFilter("chr5.fa").
> With the viral gene regions in individual fasta files, would if work to
> define cfilt <- chromosomeFilter ("Path/to/viralgene1.fa")
> and then viralgene1 <-readAligned(sp, "alignment_data.txt" filter=cfilt)?
>
> I tried to read about sfFilter, to customize a filter, but I don't see how
> you can add a path to fasta file to serve as the filter reference.
>
>
Hi, and welcome to Bioconductor.
The filtering is done based on the name of the reference sequence *in the
alignment file*. It has nothing to do with a file path. If your alignment
file has a reference sequence named FOO, then you would use that as the
filter string. Make sense?
Sean
> Thanks for any advice on how to create custom filter with fasta files as
> the
> reference for that filter.
> *
> 1.2.3 Filtering input
> Downstream analysis may often want to use a well-deï¬ned subset of reads.
> These
> can be selected with the filter argument of readAligned. There are built-in
> ï¬lters, for instance to remove all reads containing an N nucleotide, to
> select just
> those reads that map to the genome ï¬le chr5.fa, to select reads on the +
> strand,
> or to âlevel the playing ï¬eldâ by selecting only a single read for any
> chromosome,
> position and strand:
> > nfilt <- nFilter()
> > cfilt <- chromosomeFilter("chr5.fa")
> > sfilt <- strandFilter("+")
> > ofilt <- occurrenceFilter(withSread = FALSE)
> Here we select only those reads that map to chr5.fa:
> > chr5 <- readAligned(sp, "s_2_export.txt", filter = cfilt)
>
>
> # custom filter: minimum calibrated base call quality >20
> goodq <- srFilter(function(x) {
> apply(as(quality(x), "matrix"), 1, min) > 20
> }, name="GoodQualityBases")
> goodq
> aln[goodq(aln)]*
>
> [[alternative HTML version deleted]]
>
>
> _______________________________________________
> Bioc-sig-sequencing mailing list
> [email protected]
> https://stat.ethz.ch/mailman/listinfo/bioc-sig-sequencing
>
>
[[alternative HTML version deleted]]
_______________________________________________
Bioc-sig-sequencing mailing list
[email protected]
https://stat.ethz.ch/mailman/listinfo/bioc-sig-sequencing
