How To Evaluate The Effect That A Snp Has On The Structure And Function Of The Protein?

Question

[rt.cgen]

I am working on SNP and I have narrowed down my search to few SNPs which I believe is most damaging to the structure and function of the protein. I want to test this hypothesis (the hypothesis being that this particular mutation caused by SNP would be most damaging to the native protein structure) with as many ways possible. I have already tested my hypothesis by carrying out Molecular Dynamics Simulation (MDS) and Bayesian theorem and both of the fore mentioned methods corroborate my finding.

I would really appreciate if you can suggest some more ways to test my hypothesis, I am looking for computational and mathematical methods only. I know there are some servers like SIFT , Polyphen but please let me make clear that I have used them to narrow search to SNP, I now need to assert with reasonable certainty that my findings are correct.

TL;DR I am looking for methods to test a probability hypothesis.

Thanks in Advance, your help is highly appreciated.

ps - I have already asked this question here before, but did not receive any answer to my query,ant help would be really helpful

[Christof Winter]

Sounds like time to go to the wet lab to me ;-)

[rt.cgen]

I know but I need some more evidences to throw at my collaborators, before they actually get their hand dirty.

Answer 1

[João Rodrigues]

If you did MD and you did it correctly, then you have a good indicator of the changes in the stability of the molecule when compared to the wild type. Likely you won't see gigantic unfolding if you SNP is causing it, because of the limited timescale of the MD but you should see more and larger fluctuations in that area, correlated movements, etc..

Other techniques are just human. Ie. open the structure in Pymol and reason if and why the SNP will have any effect. Is it changing the polar character of a residue involved in hydrogen bonding? Is it reversing a charge? Are you changing a GLY to PHE and thus likely introducing steric hindrance to the neighbouring residues?

Computationally you have hit the ceiling..

[rt.cgen]

As I mentioned the MD analysis corroborate my findings, I can clearly see a noticeable difference in RMSF, RMSD, H bond count at particular time scale. The mutation occurs in the motor region of the protein, but I don't think so that is a reason enough for the reviewers.

Answer 2

[Chris Cole]

Have you done any sequence analysis of the protein? e.g. which residues are most conserved?, what domains does it have?, what structural features overlap with the mutated residues?

If you've done MDS, then you must have a 3D structure of the protein. Correct? Then where are the residues located in the structure? Surface or in the core? In the active site or not? At known binding interfaces with other proteins?

There are lots of questions you can ask and test computationally, but none of them will be definitive. The definitive proof will be to perform an assay on the protein that will indicative of it's function.

[rt.cgen]

I get your point and I know that to be sure I need to do experimental work, but I wanted to make sure that it is actually worth doing and so I am fishing out for as many methods as possible before that.