[Bioc-devel] cigarToRleList fails
Hervé Pagès
hpages at fhcrc.org
Thu Feb 20 23:32:48 CET 2014
Hi Jesper,
On 02/20/2014 02:13 PM, Jesper Gådin wrote:
> Very true that it is quite difficult to find the documentation when one
> is not aware of its existence :P
Yeah, this has been a source of frustration for many people. And
always a source of embarrassment (for us) when teaching our software
to beginners.
I've started to change this. In the upcoming version of BioC (2.14,
scheduled for mid-April), when you'll do ?coverage, you'll get to
choose between the 3 man pages that document coverage methods (there
is one in IRanges, one in GenomicRanges, and one in GenomicAlignments).
I want to generalize this to other generics that have methods spread
across several packages (e.g. findOverlaps, the intra- and inter-range
methods, etc...).
Having to choose between several man pages every time you do e.g.
?findOverlaps is a minor annoyance compared to not being able to
find the man page at all. (Note that if you already know where is
your favorite man page, you'll be able to direct access it with
e.g. ?GenomicRanges::findOverlaps). Nobody will ever need to use
the insane ?`findOverlaps,GenomicRanges,GenomicRanges-method` to
open that man page again. Ever! (it will still work though...)
Cheers,
H.
>
> coverage() is fast and beautiful. Thanks!
>
> /Jesper
>
>
> On Wed, Feb 19, 2014 at 9:21 PM, Hervé Pagès <hpages at fhcrc.org
> <mailto:hpages at fhcrc.org>> wrote:
>
> Hi Jesper,
>
>
> On 02/19/2014 08:44 AM, Michael Lawrence wrote:
>
> On Wed, Feb 19, 2014 at 8:39 AM, Jesper Gådin
> <jesper.gadin at gmail.com <mailto:jesper.gadin at gmail.com>>wrote:
>
> Hi Michael,
>
> Herves suggestion will probably work for my use case, but if
> there are any
> smoother ways it would be preferable.
>
> The use case is as follows:
>
> 1) calculate strand specific coverage over a region from
> GAlignments object (or file)
>
> At the moment I read a file using readGAlignmentsFromBam()
> with tag="XS",
> then filter it on "flag" and "mapq". Then I subset the
> resulting GAlignments in a minus and a plus -strand object.
> Then I calculate coverage by my own coverage function which
> uses the cigar
> information in the GAlignments object. This function is the
> one using
> cigarToRleList() at the moment. As I understand the
> coverage() function
> from the GenomicAlignments/IRanges package does not take
> into account
> cigars, or does it?
>
>
> It does take the coverage into account; specifically to exclude
> the introns
> from coverage. I think there's also an option to exclude deletions.
>
>
> Unfortunately the man page is not easy to access (you need to do
> ?`coverage,GAlignments-method`__), but it says:
>
> The methods for GAlignments and GAlignmentPairs objects do:
>
> coverage(grglist(x), ...)
>
> And if you do grglist() on a GAlignments or GAlignmentPairs objects, the
> ranges you get in the returned GRangesList object are calculated based
> on the CIGAR.
>
> Trust but verify. Here is how you can actually verify that coverage()
> does take the CIGAR into account:
>
> library(RNAseqData.HNRNPC.bam.__chr14)
> gal <- readGAlignments(RNAseqData.__HNRNPC.bam.chr14_BAMFILES[1])
> cig_op_table <- cigarOpTable(cigar(gal))
>
> First we pick up an alignment with an N in its CIGAR and do coverage()
> on it:
>
> > gal_with_N <- gal[which(cig_op_table[ , "N"] != 0)[1]]
>
> > gal_with_N
> GAlignments with 1 alignment and 0 metadata columns:
> seqnames strand cigar qwidth start end
> width
> <Rle> <Rle> <character> <integer> <integer> <integer>
> <integer>
> [1] chr14 + 55M2117N17M 72 19650072 19652260
> 2189
> ngap
> <integer>
> [1] 1
> ---
> seqlengths:
> chr1 chr10 ...
> chrY
> 249250621 135534747 ...
> 59373566
>
> > coverage(gal_with_N)$chr14
> integer-Rle of length 107349540 with 5 runs
> Lengths: 19650071 55 2117 17 87697280
> Values : 0 1 0 1 0
>
> Same thing with an alignment with an I in its CIGAR:
>
> > gal_with_I <- gal[which(cig_op_table[ , "I"] != 0)[1]]
>
> > gal_with_I
> GAlignments with 1 alignment and 0 metadata columns:
> seqnames strand cigar qwidth start end
> width
> <Rle> <Rle> <character> <integer> <integer> <integer>
> <integer>
> [1] chr14 - 64M1I7M 72 19411677 19411747
> 71
> ngap
> <integer>
> [1] 0
> ---
> seqlengths:
> chr1 chr10 ...
> chrY
> 249250621 135534747 ...
> 59373566
>
> > coverage(gal_with_I)$chr14
> integer-Rle of length 107349540 with 3 runs
> Lengths: 19411676 71 87937793 <tel:71%2087937793>
> Values : 0 1 0
>
> Same thing with an alignment with a D in its CIGAR:
>
> > gal_with_D <- gal[which(cig_op_table[ , "D"] != 0)[1]]
>
> > gal_with_D
> GAlignments with 1 alignment and 0 metadata columns:
> seqnames strand cigar qwidth start end
> width
> <Rle> <Rle> <character> <integer> <integer> <integer>
> <integer>
> [1] chr14 + 38M1D34M 72 19659063 19659135
> 73
> ngap
> <integer>
> [1] 0
> ---
> seqlengths:
> chr1 chr10 ...
> chrY
> 249250621 135534747 ...
> 59373566
>
> > coverage(gal_with_D)$chr14
> integer-Rle of length 107349540 with 3 runs
> Lengths: 19659062 73 87690405
> Values : 0 1 0
>
> Seeing is believing,
>
> Cheers,
> H.
>
>
>
> I started to look at the applyPileups() in Rsamtools which I
> can get to
> calculate coverage using cigars, but not using the strand or
> flag
> information for filtering. That solution would start from a
> bam-file
> instead of a GAlignments object, and sure I can do the
> filtering outside R.
> But it would be very convenient to do it all from within R.
>
> If there are nice solutions starting from both a GAlignments
> and a
> bam-file it would be great! =)
>
> /Jesper
>
>
>
> On Tue, Feb 18, 2014 at 10:52 PM, Michael Lawrence <
> lawrence.michael at gene.com
> <mailto:lawrence.michael at gene.com>> wrote:
>
> Hi Jesper,
>
> Would you be willing to volunteer your use case? As
> Herve hinted,
> cigarToRleList and friends are low-level helpers. There
> may be an easier
> way to achieve what you want, or an opportunity to
> improve things.
>
> Michael
>
>
> On Mon, Feb 17, 2014 at 1:10 AM, Jesper Gådin
> <jesper.gadin at gmail.com
> <mailto:jesper.gadin at gmail.com>>wrote:
>
> Hi,
>
> Have come across a cigar-vector that is problematic
> to process.
>
> #load package
>
> library(GenomicAlignments)
>
>
> #load data (see attached file)
>
> load("2014-02-17-cigarExample.__rdata")
>
>
> #run function cigarToRleList
>
> cigarRle <- cigarToRleList(cigarExample)
>
> Error in .Call2("Rle_constructor", values, lengths,
> check, 0L, PACKAGE =
> "IRanges") :
> integer overflow while summing elements in 'lengths'
>
> sessionInfo()
>
> R Under development (unstable) (2013-11-13 r64209)
> Platform: x86_64-unknown-linux-gnu (64-bit)
>
> locale:
> [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
> [3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
> [5] LC_MONETARY=en_US.UTF-8
> LC_MESSAGES=en_US.UTF-8
> [7] LC_PAPER=en_US.UTF-8 LC_NAME=C
> [9] LC_ADDRESS=C LC_TELEPHONE=C
> [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
>
> attached base packages:
> [1] parallel stats graphics grDevices utils
> datasets methods
> [8] base
>
> other attached packages:
> [1] GenomicAlignments_0.99.18 Rsamtools_1.15.26
> [3] Biostrings_2.31.12 XVector_0.3.6
> [5] GenomicRanges_1.15.26 IRanges_1.21.23
> [7] BiocGenerics_0.9.3
>
> loaded via a namespace (and not attached):
> [1] BatchJobs_1.1-1135 BBmisc_1.5
> BiocParallel_0.5.8
> bitops_1.0-6
>
> [5] brew_1.0-6 codetools_0.2-8 DBI_0.2-7
> digest_0.6.4
>
> [9] fail_1.2 foreach_1.4.1
> iterators_1.0.6 plyr_1.8
>
> [13] RSQLite_0.11.4 sendmailR_1.1-2
> stats4_3.1.0 tools_3.1.0
>
> [17] zlibbioc_1.9.0
>
> _________________________________________________
> Bioc-devel at r-project.org
> <mailto:Bioc-devel at r-project.org> mailing list
> https://stat.ethz.ch/mailman/__listinfo/bioc-devel
> <https://stat.ethz.ch/mailman/listinfo/bioc-devel>
>
>
>
>
>
> [[alternative HTML version deleted]]
>
>
>
>
> _________________________________________________
> Bioc-devel at r-project.org <mailto:Bioc-devel at r-project.org>
> mailing list
> https://stat.ethz.ch/mailman/__listinfo/bioc-devel
> <https://stat.ethz.ch/mailman/listinfo/bioc-devel>
>
>
> --
> Hervé Pagès
>
> Program in Computational Biology
> Division of Public Health Sciences
> Fred Hutchinson Cancer Research Center
> 1100 Fairview Ave. N, M1-B514
> P.O. Box 19024
> Seattle, WA 98109-1024
>
> E-mail: hpages at fhcrc.org <mailto:hpages at fhcrc.org>
> Phone: (206) 667-5791 <tel:%28206%29%20667-5791>
> Fax: (206) 667-1319 <tel:%28206%29%20667-1319>
>
>
--
Hervé Pagès
Program in Computational Biology
Division of Public Health Sciences
Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N, M1-B514
P.O. Box 19024
Seattle, WA 98109-1024
E-mail: hpages at fhcrc.org
Phone: (206) 667-5791
Fax: (206) 667-1319
More information about the Bioc-devel
mailing list