Promoter analysis - Finding potential binding and receptor site (Mar/03/2009 )
I have used the Universal Genome Walking kit to clone the promoter regions of two genes but now am not sure how to determine the potential sites.
I have found TESS and TFSEARCH but would like any other recommendations for free websites to may help.
I am also having trouble interpreting the results from these sites.
To me it makes sense that only transcription/receptor sites on the sense/top strand would be able to activate genes. Is this correct? And does that mean I can ignore all found potential binding site indicated as being on the (-) strand?
I am very confused and overwhelmed by the number of potential binding sites that I seem to have!
Well, the answer is a bit complicated. As these programs are just predictive, you can choose to believe everything or believe nothing. At the end, it has to be proven experimentally...these programs only offer you a guide of what COULD be. The software takes into account the length of element, the identity, etc and then provides a statistical result that gives and idea of the probability of being true. So, Ap1 sites will be found all over the place because the odds of finding the 4-5 nt stretch is pretty high. Does that mean that they are all true? Nope...does that mean that they are all false? Nope. On average, promoters do contain Ap1 and Sp1 sites as well as TBP, TFIID, etc, etc, but it's not an easy task to do. Also, promoters are incredibly complex and just because a result is on the (-) strand doesn't mean it's not true either.
Use the program to give you guide of what is potentially there and then you will have to prove it experimentally.
Sort of deflating, but that's what molecular biology is all about
I would look up the consensus for you protein and analyze the sequence by eye.
When looking for binding sites make deletions of your promoter and see what deletions
affect the transcription when your transfect your transcription factor into the cells.
When you narrow the region down to ~200 base pairs you can do mutagenesis and EMSA.
Have found my promoter sequence has about 90% homology with an already described promoter region in a closely related species. So this has been helpful!
I have another question too, the fragment that I obtained by Genome Walking in 5kb, pretty big compared to promoter regions of other species. If there anyway to predict the actual promoter size, ie is it possible to have cloned beyond to start of the promoter? Most other published promoters for this gene are around 2kb.
Appreciate your help,
Clone the 5 Kb region, put it into a luciferase, transfect in your transcription factor, do deletions and
see what region is necessary for response to your factor.