[BioC] [Bioc-devel] SNPscan -> SNPchip

[Rob Scharpf]

Hi Justin,

> Hi Rob/Ingo just checking out your SNPchip package and clearly  
> getting Copy
> Number Variation (CNV)and SNP calls together from genomic DNA hyb  
> data is
> crucial going forward. We and others have developed SNP/tiling  
> arrays for
> Arabidopsis and yeast and will hyb genomic DNA to genotype known  
> SNPs/CNVs
> AND identify new SFPs and CNVs.  Will your package development  
> plans include
> dual typing and identification methods?

At one level, dual typing and identification methods would presumably  
start with measurements made at the probe-level and produce an  
estimate of type and copy number for each SNP.   At a higher level, a  
common goal with high throughput SNP arrays is to find chromosomal  
features spanning 1 or more SNPs using estimates of genotype calls  
and copy number as a starting point -- estimates that are often made  
independently for each SNP.  As a first step towards the goal of  
identifying chromosomal features, we developed the SNPchip package  
with classes for organizing the data and methods for visualization  
that rely on a basic annotation of the SNPs (e.g., physical position)  
and chromosomes (e.g., chromosome size, centromere position, etc).    
A long response to a simple question, but I think SNPchip would be  
useful for visualizing the summaries produced by such algorithms, and  
as a tool for developing statistical models that use SNP-level  
estimates as a starting point.  I will post a tech report in the next  
day or two that hopefully clarifies these ideas.

> and will you be able to handle SNP
> arrays, tiling arrays, and dual SNP/tiling arrays.

You might want to check out the oligo package for this.  We'd be  
happy to work with you if the visualization tools in SNPchip do not  
work out of the box for your needs.

> Just thought I'd pass
> along the direction I see the technology going.

Thanks!  We appreciate the feedback.   Best regards-

Rob
> Justin Borevitz
>
> Date: Fri, 3 Nov 2006 07:58:00 -0500
> From: Rob Scharpf <rscharpf at jhsph.edu>
> Subject:[Bioc-devel] SNPscan -> SNPchip
> To: bioc-devel at stat.math.ethz.ch
> Message-ID: <EE052468-825B-4B88-978B-1E7A97414761 at jhsph.edu>
> Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed
>
> Hi,
>
> We changed the name of the R package SNPscan to SNPchip to avoid
> confusion with the SNPscan web-tool.
>
> SNPchip contains classes and methods for organizing and visualizing
> high throughput SNP data.  The classes are indirect extensions of
> eSet.  In particular, we define classes for copy number estimates and
> confidence scores, genotype calls and confidence scores, and a third
> class that contains both copy number and genotype calls.  This third
> class inherits methods from the other two.   We hope that SNPchip
> will provide a useful forum for continued discussion and development
> of these classes.  We will continue to work with Seth and Martin on
> the pros and cons of this setup, hopefully with feedback from other
> bioc-developers.  This feedback may also be useful for improving the
> classes defined in Biobase for SNP data.
>
> As of right now, we provide static tables for SNP- and chromosome-
> level annotation as .RData files.  Annotation packages for Affymetrix
> SNP chips need more discussion.  The wiki for the SNP annotation
> packages is still here: http://wiki.fhcrc.org/bioc/
> SNP_Annotation_Package_Discussion
>
> Thank you-
>
> Rob and Ingo
>
> rscharpf at jhsph.edu
> iruczins at jhsph.edu
>
>
>
> ------------------------------
>
> _______________________________________________
> Bioc-devel mailing list
> Bioc-devel at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioc-devel
>
>
> End of Bioc-devel Digest, Vol 32, Issue 4
> *****************************************