Hi Mark Thank you
I guess you mean during refinement (best results for me have been obtained iterating between phenix.real_space_refine and servalcat) Both programs have their routines to find a “best” weight. I did try to play around with this (in phenix using the weight parameter; in servalcat, using --weight and --no_weight_adjust), tightening stereochemistry starting from the automatically found values. But this modifies many things afterwards (especially damaging model-to-map FSCaverage or CC_mask values) The outliers are many because I have many chains, they are not really many for each chain. Perhaps of course instead of geometry idealization, a real-space sphere refine could be better in Coot. I am aware that these non-planarity issues are saying something about those local spots, but these are definitely not that clear. Isolde (from within chimerax) was extremely helpful in the last part of corrections, handling a huge complex, and more or less quickly helping me reduce significantly clashes, and most geometry issues. However, I don’t know if the amber FF has com thing weird about arginine side chains (as this is, by far, the residue that explains 90% of my planarity outliers), perhaps some artifactual issue is being introduced into the guanidium groups? (I saw something about this written by the chimerax folks) …anyway, right now trying to get the most accurate model I can. At least, I would love to see what Coot can do at trying to fix just this large set of Arg planarities Thanks for any help > On 5 Jun 2025, at 11:52 AM, White, Mark <mawh...@utmb.edu> wrote: > > Alejandro, > > It sounds like the map is over weighted in refinement, causing distortion of > the stereochemistry. Lower your RMSD target or Map Weighting to improve the > stereo chemistry in the model. At 3.7 Å a bond RMSD of 0.005 Å or less is > reasonable. > > Best regards, > Mark > Mark Andrew White, Ph.D. > Professor of Biochemistry & Molecular Biology, > Director, SCSB Macromolecular X-ray Laboratory > UTMB, Galveston, TX > 409.747.4747 > https://xray.utmb.edu <https://xray.utmb.edu/> > > From: Mailing list for users of COOT Crystallographic Software > <COOT@JISCMAIL.AC.UK <mailto:COOT@JISCMAIL.AC.UK>> on behalf of Alejandro > Buschiazzo <ale...@pasteur.edu.uy <mailto:ale...@pasteur.edu.uy>> > Sent: Thursday, June 5, 2025 7:30 AM > To: COOT@JISCMAIL.AC.UK <mailto:COOT@JISCMAIL.AC.UK> <COOT@JISCMAIL.AC.UK > <mailto:COOT@JISCMAIL.AC.UK>> > Subject: Python script to process a large list of residues for restrained > geometry idealization > > > External Email Warning: Do not click links or open attachments unless you > recognize the sender and expect the content. UTMB Email Phishing Awareness > <https://www.utmb.edu/infosec/resources/externalemail> > Dear cooters > > Could anyone help me out in efficiently addressing the following problem? > > I have a long list of residue side chains, whose planarity is to be corrected. > I am at the end of refinement (against a 3.7Å res cryoEM map), so already > happy with many other validation indices (and especially, map to model cross > correlation), so would want to change the model as little as possible. > My model is huge (many chains), which explains the long list of outliers. > > Can I feed this properly into Coot 1.1.15, such that self restrains are > calculated (let’s say using a 5Å sphere for the whole molecule; and that a > pretty tight Gemann-McClure alpha parameter is set, like 0.01-0.03), and that > then each of the listed residues gets ‘geometry-idealized’? > Given the many NCS relatives (many chains are identical in my complex), doing > this by hand is extremely slow, each time I say yes to a correction, there’s > a super long calculation (NCS matrix refinement) going through ALL chains > ….making this simple approach impracticable. > > Many thanks in advance! > Alejandro > > > -- > Alejandro Buschiazzo, PhD > Associate Professor > Laboratory of Molecular & Structural Microbiology > Institut Pasteur de Montevideo > Mataojo 2020 > Montevideo 11400 > URUGUAY > Phone: +598 25220910 ext. 120 > Fax: +598 25224185 > https://pasteur.uy/en/laboratories/molecular-and-structural-microbiology/ > > > To unsubscribe from the COOT list, click the following link: > https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=COOT&A=1 ######################################################################## 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/
