On Mon, 2021-05-03 at 17:49 +0200, Victor Tobiasson wrote: > Has anyone else encountered a refinement/regularisation issue in coot where > refinement produces a phi NaN? It seems to > be fairly new and specific to 0.9.5 and will completely freeze the coot > process. Unfortunately the only info I can > give is that it seems to be stochastic but rather uncommon and model/map > independent. While frozen coot uses very > little resources and responds to a SIGTERM command. > > Best, > Victor > > Coot version: > > 0.9.5 (revision-count 10365) > [with guile 1.8.8 embedded] > [with python 2.7.18 embedded] > Builder_info: build-host: ubuntu1604 on: Tue 16 Mar 0116.02 EDT 2021 by paule > Binary type: Linux-x86_64-ubuntu-16.04.6-python-gtk2 > > Final terminal output before freezing: > > (0) " CD2"/1/chainid="Ao"/8/PHE, occ: 1.000 with B-factor: 30.000 element: " > C" at (307.355,222.991,262.787) : 0.402 > addition of target position restraints: pre and post sizes: 5750 5751 > WARNING: observed torsion phi is a NAN! > debug-info: nan/nan > debug-info: atom indices: 3 >
Hi Victor, I have noticed this from time to time (rarely). It seems to be a result of the one of the 3 bonds involved in a torsion having a length of zero. I don't know how that could happen and it's tricky to track down (although I hope to one day). The problem goes away when I reduce the weight on the torsions or Ramachandran plot. Regards, Paul. ######################################################################## To unsubscribe from the COOT list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=COOT&A=1 This message was issued to members of www.jiscmail.ac.uk/COOT, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
