[deadally]

I am part of a research project using in silico molecular modeling to predict positions and metabolism of drugs.  Unfortunately, I do not understand all the protocol that goes into the methods of analysis

In AMBER9, we run PTraj to generate an RMSd (Root Mean Square deviation) plot (using fit) referencing the lowest energy structure, and it turns out a plot of points that represent (we assume) deviation versus time.  Our system, however, is a protein and contains nearly 8000 atoms, so is anybody able to tell me exactly what RMSd means in terms of our analysis?  Is each point an average of all the deviations of all the points with respect to the reference, or is it a specific point?

In essence, I'm looking for a more fundamental understanding of this analysis that seems to be a given in the field, but I haven't really learned the fundamentals quite yet, and I don't really know where to look either.

Thank you for the advice!

[Yggdrasil]

Your RMSD will be the average of the deviation of each atom in your structure.  However, depending on how your modeling program works, it may only use the backbone atoms to calculate the RMSD of the protein (as opposed to backbone atoms and side-chain atoms).  Wikipedia has a small article on RMSD:

http://en.wikipedia.org/wiki/RMSD

[deadally]

Thanks, Ygg

I believe AMBER does it to each atom...

So, I hink i understand correctly.  It averages all the deviations from the reference structure into one number for each structure, right?

[Yggdrasil]

Yup.