[BioC] Using GenomicRanges with table data

[Tom Oates]

Sorry for not being clear
That is indeed what I mean by "levels not matching"
So is it true that the chr1..chr2 etc as it appears in Levels will
have no effect on further analysis? even though Values & Levels for
seqnames do not match?

On Fri, Jan 11, 2013 at 1:54 PM, Steve Lianoglou
<mailinglist.honeypot at gmail.com> wrote:
> Hi
>
> On Friday, January 11, 2013, Tom Oates wrote:
>>
>> Hi
>> I am trying to use GenomicRanges as part of an anlalysis of sequencing
>> data.
>> I have a number of files which I wish to use to make GRanges objects.
>> For example:
>>
>> chr1    579578    579804    CpG_12
>> chr1    630418    630623    CpG_11
>> chr1    804552    804763    CpG_9
>> chr1    1307051    1307362    CpG_16
>> chr1    1323599    1323808    CpG_9
>> chr1    1350549    1350758    CpG_12
>> chr1    1403287    1403637    CpG_20
>> chr1    1418906    1419488    CpG_28
>>
>> This file is sorted such that chr1 is followed chr2, 3, 4, 5 etc to 20
>> (as opposed to chr10, 11...19, 2, 3 etc)
>>
>> I use to make the GRanges object
>> cpgi_gr<-GRanges(seqnames=Rle(cpgi$V1),
>> ranges=IRanges(start=cpgi$V2,end=cpgi$V3),
>> UCSC_AL_ID=cpgi$V4)
>>
>> but then if I examine
>>
>> seqnames(cpgi_gr)
>>
>>
>> I get
>> factor-Rle of length 89611 with 21 runs
>>   Lengths:  8952  5602  6133  4973  5840  4260 ...  3132  3607  2793
>> 3175  3842  1419
>>   Values :  chr1  chr2  chr3  chr4  chr5  chr6 ... chr16 chr17 chr18
>> chr19 chr20  chrX
>> Levels(21): chr1 chr10 chr11 chr12 chr13 chr14 chr15 ... chr5 chr6
>> chr7 chr8 chr9 chrX
>>
>>
>> So the Values & Levels are not matching.
>
>
> I'm not sure what you mean by this, but your object looks more or less fine
> to me.
>
> If by the "levels not matching" you mean that chr10 comes (instead of chr2)
> after chr1 in a call to "sea levels", then that should really affect your
> analysis ...
>
>
> HTH,
> Steve
>
>>   I hope to give the GRanges
>> object seqlengths of the chr lengths in the genome so I can then
>> perform flank etc tasks on the data so it is crucial that the values &
>> lengths match.  I imagine that this problem is based around my not
>> understanding either IRanges or Rle sufficiently but I have read help
>> on Rle objects & IRanges and can't work out how to ensure that the
>> formation of the GRanges object leads to the chr values matching
>>
>> Thanks
>>
>> _______________________________________________
>> Bioconductor mailing list
>> Bioconductor at r-project.org
>> https://stat.ethz.ch/mailman/listinfo/bioconductor
>> Search the archives:
>> http://news.gmane.org/gmane.science.biology.informatics.conductor
>
>
>
> --
> Steve Lianoglou
> Graduate Student: Computational Systems Biology
>  | Memorial Sloan-Kettering Cancer Center
>  | Weill Medical College of Cornell University
> Contact Info: http://cbio.mskcc.org/~lianos/contact