MASCOT results vs 2D gel image : Which one do you trust? - Protein positive hits in MASCOT do not agree with a spot's position!! (Aug/13/2008 )

This is the issue I would like to share with you and, hopefully, get some feedback. After running my MS/MS searches on MASCOT (by searching both NCBI and SwissProt), I get good scores for let's say 4 peptides. When I plot the relevant identified proteins on my 2D gel image, I notice a few weird things such as :

a. A, theoretically, 65kDa protein lies below a 61 and a 63 kDa protein!!!??
b. A protein with a pI of 5.86 lies almost directly below two other proteins with two totally aberrant pIs (5,70 and 5,18)!!!

I have noticed in a couple of publications that people do notice a sometimes big difference between the observed and the theoretical MW or pI. Is there any rationale behind all this? I know that the proteins in my gel might have modifications that MASCOT fails to disclose, but is that enough?? Any good interpretation strategy going around?? Thank you!!

Friendly yours
Ithace

I have also noticed the same thing, but I think this is not a very big issue. As your molecular weight are not too far from your mascot result. May be your proteins on 2D are modified but sometimes you can see a difference in mobility in one dimension gel also. I don't know whether its right or not but in my case I purified a protein, the expected molecular weight was 33 Kda but it usee to run above 36 Kda. I cut the band and Identified on MALDI. it was my protein only. So I dont think it makes any difference.

If your protein ID is good (ie. good score) then both results are valid, and in fact very interesting. Predicted/calculated pI, MW are exactly that, prediction/calculation according to amino acid sequence (ie. no modifications). The REAL pI and MW depends on post-tanslational modifications and truncations. Phosphorylation and glycosylations both reduce the pI and can increase apparent MW. Also truncations of proteins can reduce the MW.