On 09/10/2019 15:35, Brady Travis wrote: > I have a low-resolution EM map and have been having difficulty modeling a > 50mer piece of curved double-stranded DNA into. The global resolution is 6.5 > A and the local resolution for the DNA is lower. I started by making an ideal > B-form DNA from ‘other modeling tools.’ The map for the DNA curves slightly > throughout the length. If I try to real space refine the DNA into the map it > doesn’t work well. The base pairs don’t stay together and the DNA backbone > ends up getting distorted. I’ve tried adding base pairing restraints with > ‘Extensions > User-defined Restraints > DNA B form bond restraints’ but when > I try to real space refine after this I run into the same issues. I’m using > Coot 0.8.9.2. Is there a better way to model this?
For Cryo-EM, you should be using Coot 0.9-pre - Coot 0.8.x is a pain at low resolution/cryo-EM. Modern Coot (0.9-pre 2725429e483f002268cc8b9b6264ef7c7ee5428e or later (mid-July)) will add base pair and stacking restraints automatically. This video (for RNA) plays upon some of the themes of your message: https://www.youtube.com/watch?v=Zzim3SXnDSk I'd add a bit of Local Distance Restraints (say 4.4A) for the fragment - and that should be enough to stabilize the refinement (do an "All Atom" refine of course. Paul. p.s. rather embarrassingly, the super-coiling of the DNA model is incorrect (I haven't got around to fixing it :-(). Maybe the real-space refinement will "just fix it"... ######################################################################## To unsubscribe from the COOT list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=COOT&A=1
