How to predict effects of different aminoacids - (Apr/04/2007 )
Hi,
I don't have much expertise in proteomics or protein analysis, just the basis. After long time I have obtained a plasmid which has most of the sequence right. It is a recombinant fusion protein. In one of the proteins, there are three aminoacids differences between the published sequence and that in my plasmid. I was wondering if these aminoacids could make my protein behaves much differently to the native one. I believe that the answer is no because two of them are in the cytosolic part and the other in the extracellular side, so insertion into the plasme membrane should be possible. The two potential trasnmembrane domains don't show any difference. I think I may get away with this when I publish that data and start to transfect this construc as soon as possible. What thiings could happen due to these minor differences? Do you think the referees can say something like if it is not exactly similar then your data is not valid?
I just want a rough idea. What do you think about this? All opinions appreciated.
By the way, is there any easy way (software for instance) I can get this knowledge?
Thanks,
A puzzled graduate student
I am not expert. but i have some opinions...
It's possible to have difference of these 2 proteins. For example, if the mutation sites are important for interacting with other proteins, or these 3 aminoacides originally were charged but changed to Alanine, etc..
However, the published one maybe a mutant instead of yours.... 
OK, what residues have been changed? Where are they in the sequence? You said that you got this plasmid after a long time and a lot of work: did you make the construct yourself, or did you get it from a collaborator? If you did it all yourself, have you sequenced in both directions just to make sure there wasn't some funny sequencing artefact?
You could try putting the different sequences through an algorithm that looks at protein domains to see if they are expected to change any function. Or you could try some functional studies and see if they affect the protein.
If you are really worried about the mutations, you could try some in situ mutagenesis (but that is a fiar bit of work, so you'd want to be sure it was worth it).
Alternately, if you are confident that the differences are real, just note that there is a difference between the two constructs (yours and the published one).
Thanks for the advice. Yes, I have checked twice the sequence and probably I will do once more.
Where can I get such algorithms? Can I use them easily?
About the functional studies, I'm going to do them, but if these aminoacid differences make a big change and I can predict it theorically with such algorithms, it will save me a lot of time.
Thanks for the advice. Yes, I have checked twice the sequence and probably I will do once more.
Where can I get such algorithms? Can I use them easily?
About the functional studies, I'm going to do them, but if these aminoacid differences make a big change and I can predict it theorically with such algorithms, it will save me a lot of time.
Hi.
First of all, my apologies about the prediction algorithms, I must have been thinking of something else (must be old age finally catching up!). My first guess is that the mutations won't make too much difference as long as they are a conservative change. What organism did you use to get your gene? It could be you have a case of regular polymorphism, rather than a mutation of the "real" sequence...
There's only one way to really know the effect, and I think you can guess what that is!
there is no way you can get away with that!!!! no referee will except your conclutions cause you will never know how this mutation efect your protein/enzyme. if you had the wild type and could prouve experimentaly that this changes don't matter - that is a different thing. but now you are waisting your time! you beter go back and repair the mutations. never take the short cuts - it doesnt work in biology.