[BioC] separate MEDIPS.createSet required for each replicate? RE: using custom BSgenome with MEDIPS

[Vining, Kelly]

A follow-up question about an error message with MEDIPS.createSet command: when I load my reference genome, the create a MEDIPS set with one of my .bam files, I get the following error message:

> library(BSgenome.Ptrichocarpa.Phytozome.v3) 
> budFall2_MEDIP = MEDIPS.createSet(file="accepted_hits_F2.bam", BSgenome=BSgenome, extend=300, uniq=TRUE, shift=0, window_size=1000)
Reading bam alignment accepted_hits_F2.bam
Total number of imported short reads: 46734466
Extending reads...
Creating GRange Object...
Extract unique regions...
Number of unique short reads: 36603869
Error in as.environment(pos) :
  no item called "paste("package:", BSgenome, sep = "")" on the search list
In addition: Warning message:
In ls(paste("package:", BSgenome, sep = "")) :
  âpaste("package:", BSgenome, sep = "")â converted to character string

I don't know how to interpret this. Is this error generated from GenomicRanges? Does it indicate that I've done something incorrectly, or is it safe to ignore it?

Thanks again,
--Kelly V.
________________________________________
From: bioconductor-bounces at r-project.org [bioconductor-bounces at r-project.org] on behalf of Vining, Kelly [Kelly.Vining at oregonstate.edu]
Sent: Friday, December 20, 2013 12:34 PM
To: 'Lukas Chavez'
Cc: bioconductor at r-project.org
Subject: [BioC] separate MEDIPS.createSet required for each replicate? RE: using custom BSgenome with MEDIPS

Hi Lukas,
I have a question about the MEDIPS.createSet function with biological replicates. In the vignette, a MEDIPS.createSet command is issued for the first replicate of a set, and then other replicates are appended to make a list of MEDIPS SETs. Does that mean that one "set" equals one data set from one individual, so that I have to have a separate MEDIPS.createSet call  for each biological replicate? This seems inefficient, because the BSgenome, uniq, extend etc. parameters must be repeated for each rep even though all of those parameters are the same across all replicates.

Can I simply do something like this?

Rep1_MeDIP = system.file("extdata", rep1.bam)
Rep2_MeDIP = system.file("extdata", rep2.bam)
Rep3_MeDIP = system.file("extdata", rep3.bam)
AllReps = MEDIPS.createSet(c(Rep1_MEDIP, Rep2_MEDIP, Rep3_MEDIP, BSgenome=BSgenome)

Thanks,
--Kelly V.

From: Lukas Chavez [mailto:lukas.chavez.mailings at googlemail.com]
Sent: Wednesday, December 18, 2013 12:42 PM
To: Vining, Kelly
Cc: bioconductor at r-project.org
Subject: Re: [BioC] using custom BSgenome with MEDIPS


Hi Kelly,
let us assume that the package name of your successfully created custom BSgenome for e.g. 'oryza sativa' is 'BSgenome.OryzaSativa' and that it is installed in your R environment using R CMD INSTALL <package>.

First load your custom BSgenome object into your R session using
library(BSgenome.OryzaSativa)
Note, BSgenome.OryzaSativa will not appear in the list returned by available.genomes(). However, in case the library is loaded you should be able to immediately use it as usual in MEDIPS like

library(MEDIPS)
BSgenome=    "BSgenome.OryzaSativa"
MSet = MEDIPS.createSet(file="oryza.bam", BSgenome=BSgenome)
Lukas


On Wed, Dec 18, 2013 at 6:57 AM, Vining, Kelly <Kelly.Vining at oregonstate.edu<mailto:Kelly.Vining at oregonstate.edu>> wrote:

________________________________
From: Vining, Kelly
Sent: Tuesday, December 17, 2013 10:58 AM
To: 'Lukas Chavez'
Cc: bioconductor at r-project.org<mailto:bioconductor at r-project.org>
Subject: RE: [BioC] Question about MEDIPS extend parameter

Hello BioC group,
have created a BSgenome object for my genome of interest, and it is in my current working directory. I now want to use it with MEDIPS. How do I make it appear under
>available.genomes()? Is it necessary for my custom genome to appear in this list for me to be able to use it with MEDIPS? Or is there a different method for accessing this object?

Thanks much,
--Kelly V.



        [[alternative HTML version deleted]]

_______________________________________________
Bioconductor mailing list
Bioconductor at r-project.org<mailto:Bioconductor at r-project.org>
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor


        [[alternative HTML version deleted]]

_______________________________________________
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