Re: [ccp4bb] Going back to Coot 0.8
Dear Veronica, Sorry to re-open old thread (staycatation...) I don't think that it's old - just slow. glad for you listening to the feedback - much appreciated. Thank you, fellow in-liner. All that top posting was causing me mental anguish. I think the situation has arisen because Coot users are bringing previous experience of using RSR into their use of 0.9. Because the movement of the non-dragged atoms in 0.8.x was rather unsophisticated, many people learnt (I discovered, rather late in the day) to "flick" the dragged atom with a large and fast movement to get over a local energy barrier. (Needless to say, I hadn't intended for This +1 (everyone using coot, that I know does it) Hmmm, :-( For myself, by using eigen-flip and jiggle fit beforehand and then by combining pepflip, JED Flip and backrub rotamer during RSR I often don't even need to pull on the atoms. By Do you mean in the new version: instead of simple "drag - large and fast movement to get over a local energy barrier" - you recommend to do this above multiple steps? Sadly, it would be a great decrease in usability. That's not what I meant, yes that would be a decrease in usability. Let me have another go. Interactive real space refinement is a technique to which one turns only if faster, more automated methods fail. Initial rotation/translation means that I don't have to pull on an atom go to the right place, it just sinks in. That makes things quicker. Likewise JED flipping and pep-flipping. Regards, Paul. p.s. Anna asked a question recently about metal restraints in Refmac. In Coot, the Acedrg tables have been used to extract target distances for metals (e.g. share/coot/data/metal/metal-O-distance.table) and they are automatically applied to the current model. So one can drag on a coordinating water and the metal and the other ligands come along for the ride. This is another reason why 0.9 refinement is better. To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/
Re: [ccp4bb] Going back to Coot 0.8
Dear Paul, Sorry to re-open old thread (staycatation...) glad for you listening to the feedback - much appreciated. > I think the situation has arisen because Coot users are bringing previous > experience of using RSR into their > use of 0.9. Because the movement of the non-dragged atoms in 0.8.x was > rather unsophisticated, many people > learnt (I discovered, rather late in the day) to "flick" the dragged atom > with a large and fast movement to > get over a local energy barrier. (Needless to say, I hadn't intended for This +1 (everyone using coot, that I know does it) > For myself, by using eigen-flip and jiggle fit beforehand and then by > combining pepflip, JED Flip and > backrub rotamer during RSR I often don't even need to pull on the atoms. By Do you mean in the new version: instead of simple "drag - large and fast movement to get over a local energy barrier" - you recommend to do this above multiple steps? Sadly, it would be a great decrease in usability. Best regards, Vero PS: I still have not used the new version. > adding in interactive contact > dots, I can see what it is that's causing Coot to not move the atom to where > I'm trying to drag it (the NBCs > have been re-paramaterized and up-weighted in the 0.9.x rewrite). I have, > from time to time in the past > year, used Coot 0.8.x and to me it now feels painfully crippled. RSR in Coot > 0.9.x is joyfully expeditious > and pleasingly animated. > > For now (which is to say, before Coot 0.9.1 is available) I suggest keeping > an eye out for unsatisfied atom > pull restraints, and using the "Clear Pull Restraints" at moments of > confusion. Also, use less flick and > more smoothness when dragging atoms. For example, a 180 degree rotation of > the ribose from the Coot tutorial > would be difficult, if not impossible using 0.8.x, but in 0.9.x one can pull > on the hydrogen atom of the O5' > to rotate it in a few seconds. I have added a (rather poor) video of me > doing just that to my channel (I > should make a better video). > > With all that said, I am still listening - I will add a change shortly that > will make the pull atom > restraints more obvious by making them fatter, pinker and more opaque. If > there are still problems and you > could somehow make a screencast available to me that illustrates the > problem, then I would be very > interested to view it. > > Regards, > > Paul. > > > On 04/09/2020 10:05, Schreuder, Herman /DE wrote: >> Dear Paul, >> >> Here I fully agree with Eike. With the real space refinement in the new >> coot the ligand often goes >> everywhere, except where it should go. Changing the Xray weight helps >> sometimes, but not always. In many >> cases I do not real-space refine and leave it to Buster to do the >> refinement. It would be very good if the >> old behavior could be reinstalled. >> >> Best regards, >> >> Herman >> >> *Von:*CCP4 bulletin board *Im Auftrag von *Schulz, >> Eike-Christian >> *Gesendet:* Freitag, 4. September 2020 10:36 >> *An:* CCP4BB@JISCMAIL.AC.UK >> *Betreff:* [EXTERNAL] [ccp4bb] Going back to Coot 0.8 >> >> *EXTERNAL : *Real sender is owner-ccp...@jiscmail.ac.uk >> <mailto:owner-ccp...@jiscmail.ac.uk> >> >> Dear Paul, >> >> I have been working with coot for over 10 years now with little reason to >> complain. >> >> However, in spite of trying for a few months now, I am not getting warm >> with coot 0.9. >> >> I like the new eye-candy, and the more organized menus. But fitting >> residues and ligands into ED, has never >> before been so difficult, and frankly it annoys me that previously simple >> tasks have become an effort. It >> seems as if coot and I see different minima and we always disagree where >> to put the residue. At the moment >> all my data are at convenient resolutions of 1.7Å or better, so there is >> little ambiguity on that side. >> >> I am using all default settings, but maybe there is something that needs >> to be changed? >> >> * Is there a way to go back to the old (0.8-style) fitting functions in >> coot 0.9? If so how? >> * If not, which of the last coot versions >> >> (https://www2.mrc-lmb.cam.ac.uk/personal/pemsley/coot/binaries/release/ >> >> <https://urldefense.proofpoint.com/v2/url?u=https-3A__www2.mrc-2Dlmb.cam.ac.uk_personal_pemsley_coot_binaries_release_=DwMGaQ=Dbf9zoswcQ-CRvvI7VX5j3HvibIuT3ZiarcKl5qtMPo=HK-CY_tL8CLLA93vdywyu3qI70R4H8oHzZyRHMQu1AQ=3AflakQZ5Fz_M9sPnssJL3oU4C-u25224gN5ljs5KwA=_vveOeCJ42OXDKFFrxFP_yaBEA0xSQBIGW2RWy8kmAk=>) >> wou
Re: [ccp4bb] Going back to Coot 0.8
Not the only one Paul!! On Fri, 4 Sep 2020 at 09:37, Schulz, Eike-Christian wrote: > > > > > > > > > > > > > > > > > Dear Paul, > > > > > > I have been working with coot for over 10 years now with little reason to > complain. > > > > > > > However, in spite of trying for a few months now, I am not getting warm > with coot 0.9. > > > > > > > I like the new eye-candy, and the more organized menus. But fitting > residues and ligands into ED, has never before been so difficult, and > frankly it annoys me that previously simple tasks have become an effort. It > seems > > as if coot and I see different minima and we always disagree where to put > the residue. At the moment all my data are at convenient resolutions of > 1.7Å or better, so there is little ambiguity on that side. > > > > > > > I am using all default settings, but maybe there is something that needs > to be changed? > > > > > > > > > >- Is there a way to go back to the old (0.8-style) fitting functions >in coot 0.9? If so how? >- If not, which of the last coot versions ( >https://www2.mrc-lmb.cam.ac.uk/personal/pemsley/coot/binaries/release/) > >would you recommend? > > > > > > > With best regards, > > > > > > Eike > > > > > > > > > > > > > > > > > > > > > -- > > > To unsubscribe from the CCP4BB list, click the following link: > > > https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1 > > > > To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/
[ccp4bb] Going back to Coot 0.8
Dear Paul, I have been working with coot for over 10 years now with little reason to complain. However, in spite of trying for a few months now, I am not getting warm with coot 0.9. I like the new eye-candy, and the more organized menus. But fitting residues and ligands into ED, has never before been so difficult, and frankly it annoys me that previously simple tasks have become an effort. It seems as if coot and I see different minima and we always disagree where to put the residue. At the moment all my data are at convenient resolutions of 1.7Å or better, so there is little ambiguity on that side. I am using all default settings, but maybe there is something that needs to be changed? * Is there a way to go back to the old (0.8-style) fitting functions in coot 0.9? If so how? * If not, which of the last coot versions (https://www2.mrc-lmb.cam.ac.uk/personal/pemsley/coot/binaries/release/) would you recommend? With best regards, Eike To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/
