I've heard some people mention that this or that program isn't "up to date" with respect to the borders used in the Ramachandran plots to classify φ/ψ as being in the most favorable/acceptable/disallowed regions.
How are the accepted regions defined? What is the current accepted paper that covers this?
I wish I could answer your question outright, but this is outside my area of expertise. The best I can give you is a place to start. I expect any recent book on proteomics would cover this topic. For example Introduction to Proteomics: Principles and Applications (Methods of Biochemical Analysis) by Mishra and Blobel, Wiley 2010.
I found some promising papers with a little searching with SciFinder. Neither I nor my institution subscribe to Protein Science, but the abstract looks promising. I have skimmed the article in PNAS. Hopefully one of these articles has your answer.
Zhou, et al, "Revisiting the Ramachandran plot from a new angle", Protein Science 2011 1166-1171.
Porter and Rose, "Redrawing the Ramachandran plot after inclusion of hydrogen-bonding constraints", PNAS, *2011, 108, 1, 109-112
