David Baker in this video at minute number 15:34 talks about having a standard, easy, and routine way to design novel (synthetic) proteins that do not exist in nature. So routine that even new students can successfully and easily preform it.
Questions: Is Rosetta Design the protocol to acheive this? Is there a full document from start to finish that details this protocol? (the one in the demos details a recetpor dimer protein). From the talk I understood that an Abinitio step comes after the Design step, is there a refinement step somewhere? Is there another step that results in very low score/RMSD funnel shaped plots? Are there certain preffered starting structures? Is there any details about the probability of a successful crystal structure after a good Abinitio result?
I am looking for the details please, but any information will be valuable.
Appologies for the mass of questions, but no one arround me here is doing this work.
Thanks for the question - this is a good one! The best place to start looking for the information you seek is in peer-reviewed articles published by groups developing Rosetta algorithms. Every design problem is unique, so the protocol to create helical bundles will be different than the protocol to create rossmann-like domains, curved beta-sheet containing proteins with pockets, repeat proteins, TIM barrels, etc.
In our recent de novo peptide design paper, we included a detailed road map of our entire design procedure, including examples that you can run yourself, in the methods section and supplemental material.
I'd also like to attempt to put David's statement about design simplicity into the proper context. Think of it this way: X-ray crystallography is much more accessible today than ever before, and new students can indeed determine protein structures. However, this is generally assumed to be true only if that student is working in an X-ray crystallography lab. Protein design is similar; new students can also accomplish the design of novel proteins, but I think the reasonable assumption is that the new student is working in a protein design lab.
I'm happy to follow up if you have any specific questions about a particular design problem you may be interested in.