HI,
I am looking to do quality control on protein structures generated in Rosetta using the Rosetta Ramachadran scoring function. I was trying to understand how the Ramachadran scorer reads the ramachadran map files. From looking at the ones in the databases, such as Rama08, the first two columns correspond an amino acid type and it simply cycles through them. However which amino acid corresponds to residue type 1 versus type 2 is not clear to me. From hunting through the code it looks like the enumeration is pulled from the core::chemical::AA module. Am I correct in my understanding?
Category:
Post Situation:
Rosetta/database/scoring/score_functions/rama/Rama08.dat is actually for the two-body Ramachadran score (rama2b). The score table for the default rama term (single residue) is actually Rosetta/database/scoring/score_functions/rama/Rama_smooth_dyn.dat_ss_6.4
The format of the single residue file is
(Note that in current versions of Rosetta, the value that's used to compute energy is actually the "prob" column, rather than the "energy" column.)
And the format for the two-body file is
You are correct that the aa_num is from the enumeration in the core::chemical::AA module - specifically, it's based off the one letter code. So Alanine is 1, cystine is 2, aspartic acid is 3 ... and tryptophan is 20.