Hi Nat,
Maybe you are correct. At the original Pro position, there is almost no electronic density (out of the major electronin density body). According to the opinion of Paul from Coot, the good method would be, after I build the main peptide and do the refinemnt, I add the extra rsidues (for example the Pro we just discussed) at the terminal of the major peptide just refined according to the electronic density map exposed after refinemnt.
But here I have another question and I do not know whether you can answer. For the above method introduced by Paul, after the initial refinement of the major peptide, more electronic density map could be exposed by the refinement, and we can base the newly exposed electronic density map to rebuild the extra residues.
However for rotamer, after refinement, we change the conformation of a rotamer a little, so that the rotamer would be not a outlier. Of course in this situation, after the rotamer changes into the non-outlier, I find the non-outlier rotamer does not not fit the original electronic density map so nicely. Then I do another phenix refinemnt, I hope the electronic density may would fit to the new non-outlier rotamer. However I find the electronic density may does not change at all by further refinemnt, so it means the conformation of the non-outlier rotamer is not correct.
Will you please explain to me why in the situation of Paul, new electronin density map can occur, but for my non-outlier rotamer situation, the electronic density does not change anymore by refinemnt to suit the non-outlier rotamer?
Cheers,
Dialing
From: Nathaniel Echols
Attached is a part of electronic density pap I assigned to Pro. After Phenix refinement, we can see under the Pro ring there is a rather large green, above the Pro ring there is a part of red.
Do you think whether I assign the Pro correctly?
My guess is no, but it's difficult to be certain from a single image. If you're certain that the sequence assignment of the rest of the chain is correct (not out of register), the Pro might be a mistake (in which case you would need to have the expression construct or gene re-sequenced). I should repeat what Tim Springer told you a week or two ago: it would be wise to find a more experienced crystallographer in your department/institution (hopefully there is one available!), and have him or her go through the structure with you interactively in Coot. The bulletin boards are great resources if you need a second opinion, or require help figuring out how to use the software - however, they are not a substitute for direct assistance when you're learning out how to build a model manually. We're always happy to try to answer questions like this, of course, but I think you'll find this a much slower and less effective way to learn than talking to someone personally. -Nat _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb