Author: Brian Coventry (2020)
The SAP score is a property of proteins that determines how aggregation prone they are. This was originally developed for antibodies, but it has tons of uses for proteins. It comes from this paper.
Developability Index: A Rapid In Silico Tool for the Screening of Antibody Aggregation Propensity
Brian added various ways to calculate this as well as design with it to Rosetta. See below for the various options.
At the time of writing, this isn't published, so there's nothing to cite. I figure we keep it that way until it's important enough to someone's work that they need to describe it in the methods. At which point that'll be the publication to cite. (You could probably cite the original Developability Index paper though. The SimpleMetrics are trivial. The movers are the only place where something interesting was required.)
Someday, this documentation will be better.
Use these and you'll get values.
For these to work. You must enable the sap_constraint scoreterm. Ideally with weight = 1.
You can get whole-pose, per-residue, and per-atom sap scores with these functions.
core.pack.guidance_scoreterms.sap.calculate_sap()
core.pack.guidance_scoreterms.sap.calculate_per_res_sap()
core.pack.guidance_scoreterms.sap.calculate_per_atom_sap()