[BioC] assignChromosomeRegion in 2.6.0 ChIPpeakAnno

[Stefan Pinter]

Sure, thank you Julie. 

Yes, that is what I meant. Since all the necessary information (peak locations, feature locations, total count, total length) are already available in RangedData & Annotation, it would be very convenient to calculate enrichment/depletion and possibly even significance scores (by permutation) on the fly. The significance score may be overkill, but if the function at least reported enrichment/depletion scores, the user could always supply a number of shuffled ranges to build a random model of enrichment scores and calculate significance after.

In addition, would it be possible to add another definition for PeakLocForDistance in annotatePeakInBatch?
PeakLocForDistance = "start means using start of the peak to calculate the distance to feature"

It would be helpful to have "closest", meaning distance to feature measured from peak start or peak end, whichever is closer. That would help with broad peaks, which using "middle" for isn't very helpful.

Thank you and Best wishes,
- Stefan.

----- Original Message -----
From: "Lihua Zhu (Julie)" <Julie.Zhu at umassmed.edu>
To: "Stefan Pinter" <pinter at molbio.mgh.harvard.edu>
Cc: "<bioconductor at r-project.org>" <bioconductor at r-project.org>
Sent: Friday, March 15, 2013 8:06:03 AM
Subject: Re: [BioC] assignChromosomeRegion in 2.6.0 ChIPpeakAnno

Stefan,

You can download 2.6.1 to directly use assignChromosomeRegion without the
package prefix (I updated last night and it might take sometime to propagate
to the installation site). To find out what parameters supported by the
function, in R, please type help( ChIPpeakAnno:::assignChromosomeRegion).
You will see that indeed PeakLocForDistance is not supported. If needed, I
can add such parameter.

Also regarding your previous suggestion of adding enrichment status of
feature length, do you mean enrichment of peaks in certain category of
chromosome region? For example, a significant enrichment score with 90%
peaks in promoter region would be interesting.

Could you please keep the thread in the Bioc for others to
contribute/benefit? Thanks!

Best regards,

Julie


On 3/14/13 10:10 PM, "Stefan Pinter" <pinter at molbio.mgh.harvard.edu> wrote:

> PPS. I think PeakLocForDistance is not working in that function:
> 
>> l.feat.dRF8.100k <-ChIPpeakAnno:::assignChromosomeRegion(rda.dRF8, TSS, Exon,
>> utr5, utr3, proximal.promoter.cutoff=100000,
>> immediate.downstream.cutoff=100000, PeakLocForDistance="middle")
> Error in ChIPpeakAnno:::assignChromosomeRegion(rda.dRF8, TSS, Exon, utr5,  :
>   unused argument(s) (PeakLocForDistance = "middle")
>> l.feat.dRF8.100k <-ChIPpeakAnno:::assignChromosomeRegion(rda.dRF8, TSS, Exon,
>> utr5, utr3, proximal.promoter.cutoff=100000,
>> immediate.downstream.cutoff=100000, PeakLocForDistance= middle)
> Error in ChIPpeakAnno:::assignChromosomeRegion(rda.dRF8, TSS, Exon, utr5,  :
>   unused argument(s) (PeakLocForDistance = middle)
>> l.feat.dRF8.100k <-ChIPpeakAnno:::assignChromosomeRegion(rda.dRF8, TSS, Exon,
>> utr5, utr3, proximal.promoter.cutoff=100000,
>> immediate.downstream.cutoff=100000, PeakLocForDistance = c("middle"))
> Error in ChIPpeakAnno:::assignChromosomeRegion(rda.dRF8, TSS, Exon, utr5,  :
>   unused argument(s) (PeakLocForDistance = c("middle"))
> 
> ----- Original Message -----
> From: "Stefan Pinter" <pinter at molbio.mgh.harvard.edu>
> To: "Lihua Zhu (Julie)" <Julie.Zhu at umassmed.edu>
> Sent: Thursday, March 14, 2013 9:39:10 PM
> Subject: Re: [BioC] assignChromosomeRegion in 2.6.0 ChIPpeakAnno
> 
> PS. if I might add a simple feature addition to that function - it would be a
> TRUE/FALSE trigger for whether to also report enrichment/depletion stats based
> on feature lengths, i.e. 50% peaks labeled as intergenic (or enhancers) is
> less interesting than 50% of peaks reported as exonic (much greater enrichment
> as total exonic feature length is smaller). Thanks, Best...S
> 
> ----- Original Message -----
> From: "Stefan Pinter" <pinter at molbio.mgh.harvard.edu>
> To: "Lihua Zhu (Julie)" <Julie.Zhu at umassmed.edu>
> Cc: "bioconductor" <bioconductor at r-project.org>
> Sent: Thursday, March 14, 2013 9:32:33 PM
> Subject: Re: [BioC] assignChromosomeRegion in 2.6.0 ChIPpeakAnno
> 
> Hi Julie,
> 
> yes, that worked! Thank you for the quick help, I very much appreciate it!
> Thank you and Best wishes,
> - Stefan.
> 
> ----- Original Message -----
> From: "Lihua Zhu (Julie)" <Julie.Zhu at umassmed.edu>
> To: "Lihua Zhu (Julie)" <Julie.Zhu at umassmed.edu>, "Stefan Pinter"
> <pinter at molbio.mgh.harvard.edu>
> Cc: "<bioconductor at r-project.org>" <bioconductor at r-project.org>
> Sent: Thursday, March 14, 2013 8:50:51 PM
> Subject: Re: [BioC] assignChromosomeRegion in 2.6.0 ChIPpeakAnno
> 
> Stefan,
> 
> Thanks for reporting this! Actually the function assignChromosomeRegion  is
> not exported.
> 
> Could you please try the following to see if you can use it? Thanks!
> 
> ChIPpeakAnno:::assignChromosomeRegion
> 
> Best regards,
> 
> Julie
> 
> 
> On 3/14/13 7:36 PM, "Lihua Julie Zhu" <julie.zhu at umassmed.edu> wrote:
> 
>> Stefan,
>> 
>> Could you please send me the sessionInfo? I just want to make sure you
>> installed version 2.6.0. Thanks!
>> 
>> I noticed that when I try to install ChIPpeakAnno with the following code, I
>> get 2.4 version instead.
>> 
>> source("http://bioconductor.org/biocLite.R")
>> biocLite("ChIPpeakAnno")
>> 
>> Best regards,
>> 
>> Julie