Haynes Lab:Notebook/Synthetic Biology and Bioinformatics for Predictable Control of Therapeutic Genes/2012/02/07

From OpenWetWare
Revision as of 13:16, 8 February 2012 by Caroline Hom (talk | contribs) (More ENCODE Filtering)
Jump to: navigation, search
Owwnotebook icon.png Project name <html><img src="/images/9/94/Report.png" border="0" /></html> Main project page
<html><img src="/images/c/c3/Resultset_previous.png" border="0" /></html>Previous entry<html>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;</html>Next entry<html><img src="/images/5/5c/Resultset_next.png" border="0" /></html>

More ENCODE Filtering

  • Received a response back from ENCODE. Hopefully this procedure will create the best possible filter to fing genes with significant histone methylation peaks at the promoter

See answers interspersed below:

I am aware that I can check the boxes chrom, chromeSTART, and chromeEND and then copy and paste that into the Genome Browser. Is it possible for the tables to provide the location of the promoter (+/- 500bps to the left and right of that region) instead of a being thrown onto a random area of the gene?

The Table Browser makes it fairly easy to get regions that are some number of bases upstream of a gene; it slightly more work to get a region that is both upstream and downstream from the transcription start site. Depending on how you do it, you may wind up with gene names included or not included in your output.

There are some different options for getting a custom track that has your regions of interest with gene names. One way would be to start by getting the BED file as suggested before (be sure to include "name" in the output), and then use either your own tools (such as Excel) to add or subtract 500 bases from the appropriate lines, or use Galaxy: http://main.g2.bx.psu.edu/. Galaxy works in conjunction with the Table Browser, and it has a lot more data and text manipulation tools.

A perhaps easier way is to generate a BED file using MySQL to query the tables directly. These two queries will generate BED files from the knownCanonical and knownGene tables (knownCanonical contains one representative transcript for each cluster of transcript in UCSC Genes -- see more on the description page: http://genome.ucsc.edu/cgi-bin/hgTrackUi?db=hg19&g=knownGene):

mysql> select knownCanonical.chrom, chromStart-500 as start, chromStart+500 as end, name, 0 as score, strand from knownGene, knownCanonical where knownGene.name=knownCanonical.transcript and strand='+';

mysql> select knownCanonical.chrom, chromStart-500 as start, chromStart+500 as end, name, 0 as score, strand from knownGene, knownCanonical where knownGene.name=knownCanonical.transcript and and strand='+';

The results from these two queries can be concatenated into one file and uploaded as a custom track. Here is a session that contains exactly that:


Feel free to use the Table Browser to download the contents of the custom track . . . use either "all fields from selected table" or BED as the output format.

Also, how exactly is the histone methylation score measured? I want to be able to select a specific score range (ex:>= 500), but I am unsure of what to qualify as a significant enough of a signal since I do not know what the score is being measured relative to and what the top score possible is.

When you have the table selected in the Table Browser, hit the "describe table schema" button. You should see a description of the score, and usually, a link to the range of scores that occur in the table.

Lastly, is there a way to add the gene name to the table output? It makes it a lot easier than having to determine the gene names when some genes are very close to each other and have an area of overlap. I know that the gene name check box is available in the group knownCANONICAL, but I can't seem to find it when I am in the Regulation group with the 'selected fields from primary and related tables' output format. If you could get back to me on these items that would be great. Thank you!

Generally, when you do an intersection in the Table Browser, the fields of the table that are selected first are the fields that are retained in the output. (We don't have a way to get fields from both sets of tables, but Galaxy does.) Because you are creating a filter for the second table, you will need to first create a custom track of the regions in your second table that pass your filter. Then select your promoter custom track and intersect it with your second custom track.

  • Also a paper was found on the Hotspot scoring algorithm from 2004 by Sabo et. al

Discovery of functional noncoding elements by digital analysis of chromatin structure