[BioC] predicting transcription factors

Paul Shannon pshannon at fhcrc.org
Tue Jun 5 19:24:18 CEST 2012


Dear Jing Huang,

Let me make sure I correctly grasp your problem.

 1) You have a set of co-regulated genes 
 2) You wish to identify possible shared transcription factors for these genes

If this is an accurate statement, then one good approach is to
 
 1) Obtain promoter sequence of each gene, often estimated to be 3k upstream, and 300 bases downstream, of the transcription start site.
    BioC provides excellent tools and data for this.

 2) Search for enriched transcription factor binding sites in these promoters.

The meme website (or the downloaded meme software) is one traditional way to do the search.  We have some BioC packages, including MotIV, and my soon-to-be-released collection of transcription factor matrices, which provide a good solution for this also, and have the advantage that your analysis can be performed entirely in R, reproducibly.

Is this helpful?  I can provide more detail.

 - Paul



On Jun 5, 2012, at 9:44 AM, Jing Huang wrote:

> Hi Experts,
> 
> I am interested in predicting transcription factors for a specific family of genes. According to my readings, it is possible to predict transcription factors for the genes that are expressed accordingly with similar pattern (up or down regulated). I don't know how.
> 
> Can somebody provide advices?
> 
> Many thanks
> 
> Jing Huang PhD
> 
> OHSU
> 
> 
> 
> 	[[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



More information about the Bioconductor mailing list