Question: How to test for enrichment of three bed files that 'may' form a complex
1
gravatar for Bioradical
3.5 years ago by
Bioradical50
United States
Bioradical50 wrote:

1) I am confused about how to look for enrichment of my data. Specifically I have attempted to use phyper but I seem to misunderstand how to use the phyper function when it pertains to 'lists' of overlaps for three files. My data looks like the following:

    list a - 5,627 #list of unique peaks for a transcription factor
    list b - 2,533 #list of unique peaks for a transcription factor
    list c - 3,989 #list of unique peaks for a transcription factor

The number of common peaks (overlap) within all three files (eg: the center of a venn diagram) is 2,329. 

The total number of unique peaks possible for all three transcription factors (if they perfectly overlapped the reference data) is 8,669.

2) This data pertains to intergenic regions. I am also interested in how to test whether these transcription factors are more enriched in intergenic regions versus intragenic regions assuming I have similar data for intragenic regions.

I might simply be misunderstanding the use of the phyper function and require something completely different. Essentially I am trying to see whether these three transcription factors are enriched in the reference data region in the form of a p-value or similar.

I've begun messing around with the fisher function in bedtools. Could this be used to achieve what i'm looking for? If so, how? 

Thank you,


Carlos

ADD COMMENTlink modified 3.5 years ago by Ian5.4k • written 3.5 years ago by Bioradical50
0
gravatar for Ian
3.5 years ago by
Ian5.4k
University of Manchester, UK
Ian5.4k wrote:

Interesting question. You might be able to do this using the Genomic Hyperbrowser 'https://hyperbrowser.uio.no/hb/'.

ADD COMMENTlink written 3.5 years ago by Ian5.4k

Thanks for the comment Ian. However, your link leads me to a blank page with a "This link may not be followed from within Galaxy." error. Though I don't ever use Galaxy.

 

EDIT: Just went ahead and googled it, i'll give it a shot and then let you know what I find in another comment! Thank you!

ADD REPLYlink modified 3.5 years ago • written 3.5 years ago by Bioradical50

While I haven't fully gone through the Genomic Hyperbrowser option. I was wondering if it would be possible to take my bed file of 'complex' overlaps and then run it against a bed annotation file to test if the complex is over-represented in the genome?

ADD REPLYlink written 3.5 years ago by Bioradical50
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 618 users visited in the last hour