[Bioc-sig-seq] Accessing range values from Rle object

[Patrick Aboyoun]

P. Terry,
When in doubt, first consult the manual pages. In particular

help("RleViewsList-class")
help(viewMaxs)

For example purposes, let's use the RleList object

 > x <- RleList(a = c(0:10, 10L, 0:10), b = c(10:0, 10L, 10:0))
 > islands <- slice(x, 1)

If you want the locations of the islands, use

 > as(islands, "IRangesList")
CompressedIRangesList of length 2
$a
IRanges of length 2
     start end width
[1]     2  12    11
[2]    14  23    10

$b
IRanges of length 2
     start end width
[1]     1  10    10
[2]    12  22    11

If you want the locations of the maximums within each range, use

 > viewRangeMaxs(islands)
SimpleIRangesList of length 2
$a
IRanges of length 2
     start end width
[1]    11  12     2
[2]    23  23     1

$b
IRanges of length 2
     start end width
[1]     1   1     1
[2]    12  13     2



Patrick


On 6/21/10 10:33 AM, pterry at huskers.unl.edu wrote:
>   Dear bioc-sig-sequencing,
>
> Working through the workshop handout 'basicChipSeq.pdf' from URL:
>
> http://www.bioconductor.org/workshops/2009/SeattleNov09/ChIP-seq/BasicChipSeq.pdf
>
> page 3, I would like to access range information for a subset of the records in an Rle object produced by for example,
>
>    
>> cov.ctcf<- coverage(ctcf, width = mouse.chromlens, extend = 126L)
>> cov.ctcf$chr10
>>      
> Then if run
>
>    
>> islands<- slice(cov.ctcf$chr10, lower = 1)
>> islands
>> viewMaxs(head(islands))
>>      
> would like to access subset of ranges associated with particular values in the ViewMaxs(islands) output, for example,>= to a specified depth of reads.
>
> Idea: if islands in 'islands' object are in same order as their associated depth values in viewMaxs(islands) output vector, perhaps could loop through viewMaxs(islands) vector of values, access/extract associated range from the 'islands' (Rle) object?
>
> However, how to 'access' the range values from the 'islands' Rle object?
>
> Perhaps there is a recomended way to achieve this goal?
>
>
> Thanks,
> P. Terry
> pterry at huskers.unl.edu
>
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> Bioc-sig-sequencing mailing list
> Bioc-sig-sequencing at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioc-sig-sequencing
>