Skip to content

[Bioc-devel] Fast accessor for assay names of SummarizedExperiment

2 messages · Peter Hickey, Martin Morgan

Original
Peter Hickey
# 
Hi Martin,

Accessing the names of the assays in a SummarizedExperiment object is very slow when doing this by what seems that natural way of names(assay(se)). I now see that it should be done by names(assay(se, withDimnames = FALSE). But I was only able to figure this out by some googling which led me to https://stat.ethz.ch/pipermail/bioc-devel/2014-May/005724.html - perhaps it's worth including some documentation on the implications of 'withDimnames' when accessing the assays slot (I still don't fully understand it).

I also noticed that Kasper implemented an assayNames() helper function in his bsseq package that is ever so slightly faster. Do others think a general assayNames() helper function is worth including in GenomicRanges?

Benchmarks at https://gist.github.com/PeteHaitch/346b924e3eca8c3cfade

Thanks,
Pete

 

--------------------------------
Peter Hickey,
PhD Student/Research Assistant,
Bioinformatics Division,
Walter and Eliza Hall Institute of Medical Research,
1G Royal Parade, Parkville, Vic 3052, Australia.
Ph: +613 9345 2324

hickey at wehi.edu.au
http://www.wehi.edu.au

______________________________________________________________________
The information in this email is confidential and intend...{{dropped:6}}
Martin Morgan
#
On 11/25/2014 04:46 PM, Peter Hickey wrote:
Hi Pete -- I added assayNames and assayNames<- to GenomicRanges, and updated 
bsseq (hope that's ok, Kasper?) to implement a method on the generic; my 
implementation does not use direct slot / field access, so is a lot faster but 
not the fastest.

I also documented the consequences of withDimnames=FALSE.

Thanks!

Martin