To improve the clashscore you could try increasing the value of the nonbonded_weight parameter, starting at 1000 (which I understand at one time was the default in phenix.refine, but recently for me already gives a huge decrease in clashscore). Since this will be avoiding clashes at the expense of (a) not fitting the map as well, and (b) potentially not fitting standard geometry as well, you need to watch that it doesn't increase real-space R or RMSD bonds/angles significantly. (My experience is with reciprocal space phenix.refine, but I expect there is a similar parameter in real-space_refine) First look for errors in the model that may cause classhes- more the .geo file and search to "Nonbonded". The interactions are listed under this, starting with the worst offenders. Examine those residues and see if thay are right. But if you say you have fixed these and the refinement reverts to bad values because map quality is poor, I think nonbonded_weight is the key. eab
Srivastava, Dhiraj wrote on 7/25/2024 10:08 AM:
Hi We solved several structures of a protein (in different ligand bound form) using cryoEM. While I was able to get reasonable fewer clash score and rotamer outlier in my best data set, in other data set with poor map quality, I am getting lots of clashes and rotamers outlier. I am fixing these issues and after refining the structure using phenix real space refinement (global minimization, local grid search and adp), it's getting back to same old structure with clashes. The reason is poor quality map. does anyone know a way to fix this issue? Thank you Dhiraj ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
######################################################################## To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1 This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
