Well - I would look at the deviants in COOT and see if there is a proper reason for the angels - water molecules too close? alternate conformation for some near by sidechain? Any refinement program wants to get good geometry unless there is an obstacle. If there is nothing obvious to fix you could use COOT to regulise the geometry for those residues and thus improve the model..
Good luck Eleanor On Fri, 8 Apr 2022 at 13:36, Pallan, Pradeep S < 000078fe4ea98718-dmarc-requ...@jiscmail.ac.uk> wrote: > REFMAC5 refinement: nucleic acid residues with bad geometry > > Hi All, > I am in the final refinement stage of an X-ray structure of a protein-DNA > complex, > 2.3 A resolution, using Refmac5 (REFMAC 5.8.0267, CCP4Interface 7.1.018, > Linux platform). > I am confident about the space group, refinement steps, dictionary files > and such. > > In preparation for PDB deposition, I did a validation check and found that > there are 12 instances > where the angles of DNA res deviate. My question is how can I improve the > angles > in my refinement so that it's close to ideal geometry? In Refmac > refinement, > "geometric restraints/geometric parameters", what is the range of 'Angle > overall-wt' > that one can use? I have tried varying the 'wt' (upto 2.5) and found that > the angles > have improved geometry, while a few remained. > > > Expected bond angle (°) Measured bond angle (°) > 1/D 3/DC/C1' 3/DC/O4' 3/DC/C4' 110.1 ± 1.0 > 103.9 > 1/D 5/DA/OP1 5/DA/P 5/DA/OP2 119.6 ± 1.5 128.6 > 1/D 5/DA/O5' 5/DA/P 5/DA/OP2 105.7 ± 0.9 > 99.2 > 1/D 5/DA/C3' 5/DA/O3' 6/DA/P 119.7 ± 1.2 > 112.3 > 1/D 7/DT/O4' 7/DT/C1' 7/DT/N1 108.3 ± 0.3 > 110.3 > 1/D 7/DT/C3' 7/DT/O3' 8/DT/P 119.7 ± 1.2 > 111.1 > 1/D 9/EX/C3' 9/EX/O3' 10/DG/P 119.7 ± 1.2 > 112.3 > 1/E 1/DC/C3' 1/DC/O3' 2/DG/P 119.7 ± 1.2 > 110.5 > 1/E 4/6OG/C3' 4/6OG/O3' 5/DA/P 119.7 ± 1.2 > 111.8 > 1/E 5/DA/O5' 5/DA/P 5/DA/OP1 105.7 ± 0.9 > 98.2 > 1/E 6/DA/C1' 6/DA/O4' 6/DA/C4' 110.1 ± 1.0 > 103.8 > 1/E 6/DA/O4' 6/DA/C1' 6/DA/N9 108.3 ± 0.3 > 110.3 > > As a test, I have tried refining the same structure in Phenix.refine, and > the overall > geometry was better, but still have 4 angle violations. I do not want to > use phenix > in this case, as my refinement was carried out using Refmac. > > Any thoughts? With regards to Refmac refinement, the geometry issues are > common for RNA/DNA refinements. Please correct me if I am wrong. In my > specific case, > it's a protein-DNA complex with about 410 amino acid and 24 nucleic acid > residues, and no angle deviations for protein residues! > > Thanks, > Pradeep Pallan > > ------------------------------ > > 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/
