ChIP-Seq TFBS Analysis

[maneeshi]

Hi all,
I am hoping to get some help with TFBS analysis with ChIP-Seq data. I have done de novo motif analysis suing MEME, R-SAT, but now I want to know what are the other predicted TFBS in my ChIP_Seq peaks.
I have done a lot of TFBS analysis over years with 1-10 sequences at a time. But with ChIP-Seq you have thousands of peak sequences and most tools online do not accept that much sequence. So I tried the BIOBASE TRANSFAC tool for TFBS analysis on ChIP-Seq data, it worked great, but for once in a while use its too expensive to buy
Does any one know how to download the public version of TRANSFAC database or even JASPER TFBS matrix files? I have been struggling to get these. But I know from different publications that people have used them with their own homemade scripts.
Also does anyone know of a program or script that you can use with these TFBS matrix files to look for TFBS in ChIP-Seq data?
(I am assuming this will be all based on PWM thresholds).

I appreciate any help I can get with this!

Thank you!

[sowmyai]

You might want to check out FIMO in the MEME suite of tools.

http://meme.nbcr.net/meme/fimo-intro.html

[maneeshi]

You might want to check out FIMO in the MEME suite of tools.

http://meme.nbcr.net/meme/fimo-intro.html

Thanks for your response!

However, FIMO also requires us to provide the specific motifs.
'FIMO searches a sequence database for occurrences of user provided motifs, treating each motif independently.'

My problem is that I want to know what are all the TFBS in my ChIP-Seq peaks from all the known TFBS matrices. Just like when you take a 500bp sequence for a gene promoter and analyze for any known TFBS. I want to do that for whole ChIP-Seq dataset.
So for this I need the file with all the TFBS matrices, and as FIMO mentions, I do need to provide a list of motifs that it can look for.
Does any one have this TRANSFAC or JASPER TFBS matrix file?

Thanks you!

[crazyhottommy]

try pscan CHIP http://159.149.160.51/pscan_chip_dev/

or http://trap.molgen.mpg.de/cgi-bin/home.cgi

[sowmyai]

What about the JASPAR web site? There is a "Download" button that should give you exactly the file you are looking for (unless I am misunderstanding)

http://jaspar.genereg.net

PFMs are in

http://jaspar.genereg.net/html/DOWNLOAD/JASPAR_CORE/pfm/nonredundant/

[maneeshi]

What about the JASPAR web site? There is a "Download" button that should give you exactly the file you are looking for (unless I am misunderstanding)

http://jaspar.genereg.net

PFMs are in
http://jaspar.genereg.net/html/DOWNL.../nonredundant/

Thank you so much! yes, I had clicked on the download link but could not figure out which set to download for vertebrates. Anyways, your second direct download link was useful, got the folder.
Again thank you so much!
Now I still need the TRANSFAC public database link to complete the whole set of TFBS, since JASPER coverage is a little less, but combined with TRANSFAC public it covers most of the known TFBS.

[maneeshi]

try pscan CHIP http://159.149.160.51/pscan_chip_dev/

or http://trap.molgen.mpg.de/cgi-bin/home.cgi

Thank you! Both links are helpful, but pScan is only for mouse and humans, I work in frogs
I am running TRAP now, but it only takes unto 3000 sequences, almost 10-20 times less than the ChIP_Seq data. But if it works I am still willing to upload my sequences 20 times.
One good thing about TRAP is that it uses TRANSFAC database 10.2, so many more known TFBS matrices are in it. Not sure what the output will be since it looks for enriched sites only. I will post another reply based on how well it worked.

[maneeshi]

try pscan CHIP http://159.149.160.51/pscan_chip_dev/

or http://trap.molgen.mpg.de/cgi-bin/home.cgi

TRAP: This program never finished the analysis. I kept it running over the weekend but it never gave any output. I tried it with one or two peak sequences and it worked. I am guessing the whole ChIP-Seq data is too much to handle for it.

[Cygnus]

Hi,

you might want to try the R/Bioconductor package TFBSTool (http://www.bioconductor.org/packages...TFBSTools.html)
You can directly get the JASPAR matrices you want and it works great even on big ChIP-seq data sets. So you don't need Transfac and you don't need to download the JASPAR database yourselfe.