[BioC] Detecting overrepresented genomic loci

[Furge, Kyle]

We have recently put up a package called "ideogram" that assists in plotting
data in a genomic context.  The package may be of help in your analysis and
can be found at 
http://www.vai.org/vari/supplemental/kyle.furge/download.html. The package
also includes a function named (for better or worse) "ideograb" to
semi-interactively select sets of adjacent genes for further analysis.

While the previously described two-table testing methods may be more
appropriate for your particular analysis, in addition to aCGH, GLAD, etc.
you also may want to consider the CLAC package.

Also, be aware that if a cluster of differentially expressed genes is biased
towards one sample group, this often indicts that a common underlying
cytogenetic event has taken place (i.e. DNA copy number gain / loss). The
dependence of gene expression on DNA copy number can either confound or
confirm your regional discriminate analysis depending on your experimental
design.

-kyle 


> From: jeffrey rasmussen <rasmuss at u.washington.edu>
> Date: Tue, 29 Mar 2005 13:07:49 -0800 (PST)
> To: bioconductor at stat.math.ethz.ch
> Subject: [BioC] Detecting overrepresented genomic loci
> 
> Hi all,
> 
> My goal is to discover regions with multiple adjacent genes coordinately
> regulated. Are there any packages that would allow one to test for genomic
> loci that are overrepresented in a subset of genes versus the remaining
> genes on the chip? I have tried using the geneplotter package to visualize
> the genomic distribution of differentially expressed genes, however, I
> have found it difficult to identify candidate loci from these plots
> because neighboring genes are indistinguishable from each other on the
> output graphic.
> 
> Thanks,
> Jeff.
> 
> __________________________________
> Jeffrey Rasmussen
> Research Consultant, Bioinformatics
> Department of Immunology
> University of Washington
> 
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> 


This email message, including any attachments, is for the so...{{dropped}}