I think doing Co-IP will sort out your problem !!!

Using Co-IP is not really usefull for this case as Pankaj doesn't actually know what protein he is looking for (did I get it rigth?) and whants to screen for possible TF that bind to a region.

A website that I personally use for predicting TF binding site is ECRbrowser. on this website you can select your gene and all the sequence will be given to you, then you have to chose a part of the sequence within the evolutionary conserved ones (usually higligtehd by a coloured bar), at this point you can ask to blast this sequence with the consensus sequence of many known TF or customs one.

http://ecrbrowser.dcode.org/

Thank You guys...

Dear Elia, you are right I don't know the transcription factor I just know the gene sequence.

Then you can predict the binding site with those tools. Then if you want to confirm some of the TF you will have found you can perform a chromatin immuno recipitation, provided you have a good antibody for the TF you will be interested to confirm.

You are welcome

Pankaj- the best choice for approach may depend on the type of organism and cell/tissue type you want to study. Can you please elaborate on these parameters?

Dear Daniel,

I want to find out the transcription factor that is regulating my gene of interest in rat heart tissue.

Pankaj- species info is highly relevant. ECRbrowser will let you align evolutionarily conserved motifs against the rat genome. However, I usually prefer to base my analyses off of ChIP-Seq datasets, such as those provided by ENCODE. Unfortunately, with rat as the target genome, the utility of this approach is limited. Still, you might look for other publically deposited ChIP-Seq dataset for rat, and/or make some inferences based on the human or mouse genomics datasets. ChIP-Seq datasets can often yield better insights since 1) the presence of a particular motif does not prove that the motif is functional, and 2) many motifs are recognized by a large family of transcription factors, limiting the ability to infer which family member(s) is/are important to your system and GOI.

Even though there is a lot of databases available which contain the putative binding site for numerous TFs, I have found it more useful searching for these binding sites based on evolutionary conservation. That way you have the chance to find sites which are potentially functional and relevant, since they have been conserved, even if these sites have not been reported previously. The same applies to co-regulated genes, which are subjected to some common regulatory mechanism.

If you can get the promoter region of orthologous genes or genes co-regulated with your gene of interest, you can try tools such as Melina (http://www.openhelix.com/cgi/tutorialInfo.cgi?id=26) or SCOPE (http://genie.dartmouth.edu/scope/) to find relevant regulatory motifs.

Hi Pankaj, I agree with Gustavo in that you should see if conservation can tell you something. I'm recommending some tools here: You can extract orthologous promoter regions using ENSEMBL's Biomart, then align them using a promoter-alignment tool like Pro-Coffee (can be run on the T-Coffee server). Then you can try to scan your alignment for known motifs using position-specific scoring matrices. For this I would use MotEvo (recently published in Bioinformatics), it's downloadable from the SwissRegulon data base and comes along with a set of weight matrices. It also tells you where unknown but highly conserved motifs are sitting.

Dear Pankaj,

There is a technique called "reverse ChIP". You can use an oligo complementary to the promoter sequence of choice and pull down all proteins associated with your promoter in vivo. Then, all proteins (hopefully, the TF you are hunting for..) can be identified by proteomic approaches.

You can read more here (http://www.nature.com/nmeth/journal/v6/n3/full/nmeth0309-187.html)

We can discuss more, if you come down to CCMB sometime......

Sure- reverse ChIP is great for highly repetitive sequences that ensure high yields of DNA for the mass spec studies. But can this realistically be applied to a promoter that is only present at a diploid copy number of 2? Yatendra- are you able to do this? From how many starting cells as input?

@Danial

I have not tried it yet. I am going to do it on yeast cells so cell material should not be a problem.

However, I am scanning the articles citing Kingston's paper to find out more about it.....

False positives is the major problem in predicting TFBSs in a given sequence of interest. Thus, any computer program that employs the phylogenetic foorprinting would be the best for such task.

I would also first go for Daniels suggestion to explore experimental binding data, if available. Maybe a genome browser (e.g. from UCSC) is useful, so you can easily inspect the data by eye. But be aware that even in vivo, binding of a transcription factor is not always functional. In that respect, it is also helpful to check whether possible binding sites are conserved.

Construct an dna fragment with the sequence of interest, immobilise it to media (sepharose bead / paramagnetic bead) incubate bead in a cell extract, wash beads in increasing salt concs, use eluted protein for 2d gel and send for mass spec

To identify putative TF binding sites the MEME suite could be an option (with tools like FIMO to identify matches of PWMs for a TF in the target nucleotide sequences). Another good option is HOMER, which can be used for sequences or peak locations from ChIP-seq data (if you happen to have it for another TF and you want to know the putative TFs that may be co-binding or binding close). There are many other tools, but these are the ones I use more. Beware of false positives!

There is a suit of tools in Genomatix for TF binding sites search. It also takes into account evolutionary info as it was suggested above. Genomatix is not free, but you can get a trial account if you need one time job to be done.

Thank you all for your valuable suggestions and information...

Have a nice time....

Hi, I highly recommend Genomatix software, it is an excellent tool because you can use it for different kinds of analysis such as: NGS, Methylation and of course to identify transcription factors in a specific sequence ( MatInspector tool). However Genomatix is not for free, you must pay a 1-year license, but it is a great software.

http://www.genomatix.de/

Use ENCODE. Its very helpful

We have recently a new software pipeline called Haystack find TF enriched :

https://github.com/lucapinello/Haystack/blob/master/README.md

It also integrate gene expression data and epigenetic data if you have.

We validated the pipeline in this PNAS paper:

http://www.pnas.org/content/111/3/E344.abstract

Any feedback is well appreciated!

No-one has suggested a yeast one-hybrid? It is a yeast-based assay that uses your DNA sequence of interest as bait to "fish" for interacting transcription factors. A nice summary here: http://bitesizebio.com/25900/an-overview-of-the-yeast-one-hybrid-assay/

But it's widely cited: https://scholar.google.com.au/scholar?hl=en&q=yeast+one+hybrid&btnG=&as_sdt=1%2C5&as_sdtp=

I'm just wondering if there is a reason it hasn't been suggested here - maybe there is a reason these researchers don't use it?

Is there any online tool suitable to determine the possible binding site of a TF in fission yeast?

Hi Dr. Pankaj

You can read this reference1

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3080775/

2

https://www.protocol-online.org/biology-forums/posts/30109.html

3

https://www.pnas.org/content/100/9/5313.full.pdf

4

https://molbiol-tools.ca/Transcriptional_factors.htm

5

https://nar.oxfordjournals.org/content/34/suppl_1/D74.full

6

https://www.bioconductor.org/help/workflows/generegulation/

There are many methods to isolate sequence-specific binding proteins. Some are listed as follows.

nucleic acid affinity purification

Y1H

electrophoretic mobility shift assays (EMSAs)

DNA sampling

PICh and nucleic acid affinity purification with SILAC (collectively known as reverse ChIP)

iChIP

ChAP-MS

TAL-ChAP-MS

TChP

CRISPR-ChAP-MS

CasID

CAPTURE (CRISPR affinity purificationin situ of regulatory elements)

In theory, all types of experiments are wonderful. BUT, in practice, anyone of them doesn't work perfect. I tried some of them in years, I have still no specific protein.

Since you have no idea, which TF actually binds to your promoter, I would suggest to predict a list of TF. There are some tools, which can do that for you, but most of them need a binding matrix. Unfortunately, most of the freely available matrices data banks are not very good/accurate/complete.

We have quite some experience in our lab doing these kind of analysis. I can offer you to run a quick analysis, if you send me a FASTA file of the promoter region. We usually consider the promoter to be 1kb upstream and 100bp downstream of the transcription start site, but you can add a bit more to make sure to not miss something ...