On 09/09/2020 11:37, Eleanor Dodson wrote:
I know how difficult it is to get the chemistry right, and that really has to be up to the researcher.. a database of high resolution structures with metals would be useful as a sanity check. Robbie mentionedMetalPDB (http://metalweb.cerm.unifi.it/). EOn Wed, 9 Sep 2020 at 11:18, Garib Murshudov <ga...@mrc-lmb.cam.ac.uk <mailto:ga...@mrc-lmb.cam.ac.uk>> wrote:Hi Eleanor, Obviously I agree that people should use restraints for metals. It will make metal refinement easy and coordination consistent with chemistry For distance and angle restraints there is an external distance restraint mechanism. PDB-REDO can generate for some of the metals. Unfortunately it is not easy in general. But can be done manually. If you know metal and coordinating atoms and coordination then putting restraints manually in ccp4i2 refmac keyword part should be straightforward. For example (I think it is correct) External distance first chain A residue 239 atom ZN second chain A residue 500 atom O value 1.95 sigma 0.032 type 0 External distance first chain A residue 239 atom ZN second chain A residue 145 atom ND1 value 2.03 sigma 0.05 type 0 External angle first chain A residue 500 atom O next chain A residue 239 atom ZN next chain residue 145 atom ND1 value 109 sigma 5 type 0 Unfortunately these restraints do not account for the fact that when Zn is bound to O or N atoms then it may change the nature of bonds (i.e. ZN bonds are often covalent like, as Robbie says: these are d-block atoms for you) The values above assume that ZN coordination is 4 and it is tetrahedral. For others similar restraints could be added. In future we would like to semi-automate metal refinement. At the moment it is pretty much users’ responsibility to design correct restraints (it is not ideal, not because we do not trust users but because the knowledge is not easily transferable from one refinement to another). Regards GaribOn 8 Sep 2020, at 20:51, Eleanor Dodson <eleanor.dod...@york.ac.uk <mailto:eleanor.dod...@york.ac.uk>> wrote: Yes Garib - all true but I dont think people should try to impose restraints initially in refinement of metals. I never (knowingly) keep the metals in my MR model. Searching for them with phases from a putative MR solution is one of the best verifications that it is right... You can use anomalous fouriers to fix the metals as accurately as possible. That could help with issue 3. And peak heights give you a bit of information re occupancy. Surely issue 2) shouldnt happen! Do you mean BUCCANEER builds the sequence wrongly? Na etc.. Impossible - hate them.. 4) & 5) are more or less the same .. occupancy is tricky - again anom peaks ? Occupancy refinement? all difficult at low resolution.. On Tue, 8 Sep 2020 at 20:29, Garib Murshudov <ga...@mrc-lmb.cam.ac.uk <mailto:ga...@mrc-lmb.cam.ac.uk>> wrote: Hi Jan, It is my experience also that if atoms are in more or less correct positions then non-bonding interaction together keep metals in correct positions, perhaps with a little bit different metal-coordinating atom distances. Problems arise when 1) one or several of the corrdinating atoms are missing which is the case at low resolution. Light atoms are almost invisible around heavy atoms (series termination and B value effects); 2) there are conflicting restraints which often happens with SS-bond (wrong SS-bond) and Zn-Cys bonds; 3) Starting atomic coordinates are far from right positions (which happens after molecular replacement when conformations are not exactly same around metals; 4) Metals are not fully occupied (which happens when metals involve reaction and they arrive during or as a part of reaction); 5) non-specific metal binding sites (which happens when metals are not part of the molecule but are used as a part of sample preparation - crystallisation or otherwise). 6) light metals (Na and similar) when metals are difficult to distinguish from water molecules; 7) highly mobile metals. Regards GaribOn 8 Sep 2020, at 13:01, Jan Dohnalek <dohnalek...@gmail.com <mailto:dohnalek...@gmail.com>> wrote: Hi Garib,On 8 Sep 2020, at 11:39, Jan Dohnalek <dohnalek...@gmail.com <mailto:dohnalek...@gmail.com>> wrote: These are structural.Are they tetrahedral or octahedral? From the list of neighbours they do not look like tetrahedral. Some of them do look like octahedral. They are involved in reaction. Two are ~ octahedral (skewed though, two positions filled by catalysis participant), one is ~tetrahedral, but actually can also accept a fifth coordinating atom. But as I said - in all our structures restraining the coordination geometry is not necessary, they hold nice. JanJan On Tue, Sep 8, 2020 at 12:22 PM Garib Murshudov <ga...@mrc-lmb.cam.ac.uk <mailto:ga...@mrc-lmb.cam.ac.uk>> wrote: What are these numbers? If I understand these numbers correctly: none of your Zn atoms is structural (4 coordinated tetrahedral). If that is the case then you need specific links or restraints. If my reading of your numbers is correct then there could be some chemistry change of the surrounding residues. If it is not structural Zn then it is likely that coordination is 6. But without seeing coordinates and maps it is difficult to say what is there. Regards GaribOn 8 Sep 2020, at 11:11, Eleanor Dodson <eleanor.dod...@york.ac.uk <mailto:eleanor.dod...@york.ac.uk>> wrote: Hmm - here is my problem - a list of ZN contacts for the two molecules.. residue 602 is a phosphate, and there possibly should be a few more waters .. No idea how best to tackle it.. E Z401 ZN A W 21 NA2.057 X,Y,Z 1.008.73 Z401 ZN A W 21 OA2.220 X,Y,Z 1.008.76 Z401 ZN A H 26 NE2A2.000 X,Y,Z 1.008.39 Z401 ZN A D139 OD1A2.085 X,Y,Z 1.008.61 Z401 ZN A Z601 O2 A1.927 X,Y,Z 0.60 10.74 Z401 ZN A O821 OC2.006 X,Y,Z 0.407.51 Z402 ZN A H 80 ND1A2.033 X,Y,Z 1.008.94 Z402 ZN A H135 NE2A2.032 X,Y,Z 1.008.70 Z402 ZN A D139 OD2A2.024 X,Y,Z 1.008.70 Z402 ZN A Z601 O2 A2.131 X,Y,Z 0.60 11.05 Z402 ZN A O821 OC1.829 X,Y,Z 0.407.81 Z403 ZN A H145 NE2A2.027 X,Y,Z 1.00 10.50 Z403 ZN A H168 NE2A2.030 X,Y,Z 1.00 10.19 Z403 ZN A D172 OD2A2.062 X,Y,Z 1.00 12.66 Z403 ZN A Z601 O3 A1.953 X,Y,Z 0.60 11.54 Z403 ZN A O820 OC2.207 X,Y,Z 0.209.09 Z403 ZN A O822 OC2.059 X,Y,Z 0.40 13.79 Z401 ZN A Z402 ZN A3.349 X,Y,Z 1.008.73 Z401 ZN B W 21 NB2.099 X,Y,Z 1.009.22 Z401 ZN B W 21 OB2.184 X,Y,Z 1.008.91 Z401 ZN B H 26 NE2B2.009 X,Y,Z 1.008.79 Z401 ZN B D139 OD1B2.069 X,Y,Z 1.008.76 Z401 ZN B Z601 O3 B1.981 X,Y,Z 0.709.31 Z402 ZN B H 80 ND1B2.032 X,Y,Z 1.009.49 Z402 ZN B H135 NE2B2.024 X,Y,Z 1.009.22 Z402 ZN B D139 OD2B2.032 X,Y,Z 1.009.70 Z402 ZN B Z601 O3 B1.973 X,Y,Z 0.709.58 Z403 ZN B H145 NE2B2.027 X,Y,Z 1.00 10.80 Z403 ZN B H168 NE2B2.029 X,Y,Z 1.00 10.65 Z403 ZN B D172 OD2B2.089 X,Y,Z 1.00 13.12 Z403 ZN B Z601 O4 B1.938 X,Y,Z 0.70 14.10 Z403 ZN B O825 OC2.322 X,Y,Z 0.20 10.61 ~ On Tue, 8 Sep 2020 at 10:47, Garib Murshudov <ga...@mrc-lmb.cam.ac.uk <mailto:ga...@mrc-lmb.cam.ac.uk>> wrote: Hi Robbie and Eleanor There are links for Zn-His and Zn-Cys. They meant to be used automatically, obviously something is not entirely right. Link names are: ZN-CYS It has a bond between Zn and S as well as an angle: ZN-CYS 1 ZN 2 SG 2 CB 109.000 3.000 This also removes H of Cys to make covalent bond between Zn and Cys. Similar links are available for Zn and His ND1 and Zn - HIS NE2 Link names are: ZN-HISND ZN-HISNE Again these links have angles between Zn and atoms of His. Angle centred at Zn is missing. But these distances and angles defined in the link it should work fine. Regards GaribOn 8 Sep 2020, at 10:40, Robbie Joosten <robbie_joos...@hotmail.com <mailto:robbie_joos...@hotmail.com>> wrote: Hi Elanor, The distances are in the dictionaries but the angles involve three different residues so these cannot be in the current dictionary. We could add the program that generates these restraints to CCP4 though. Cheers, Robbie-----Original Message----- From: Eleanor Dodson <eleanor.dod...@york.ac.uk <mailto:eleanor.dod...@york.ac.uk>> Sent: Tuesday, September 8, 2020 11:38 To: Robbie Joosten <robbie_joos...@hotmail.com <mailto:robbie_joos...@hotmail.com>>; Garib N Murshudov <ga...@mrc-lmb.cam.ac.uk <mailto:ga...@mrc-lmb.cam.ac.uk>> Cc: CCP4BB@JISCMAIL.AC.UK <mailto:CCP4BB@JISCMAIL.AC.UK>; Robert Nicholls <nicholls@mrc- lmb.cam.ac.uk <http://lmb.cam.ac.uk/>> Subject: Re: [ccp4bb] metal coordination at low resolution - restraints Robbie - could that be added to the distributed dictionaries? Zn binding is common and at low resolution distance restraints are not enough.. Eleanor On Tue, 8 Sep 2020 at 10:33, Robbie Joosten <robbie_joos...@hotmail.com <mailto:robbie_joos...@hotmail.com> <mailto:robbie_joos...@hotmail.com> > wrote: Hi Anna, Yes you can do this in Refmac by adding external restraints. If you have structural Zinc sites (Zn coordinated by 4 histidines or cysteines) you can also use PDB-REDO to generate the restraints automatically. The restraints are written to the output so you can continue using them in Refmac. HTH, Robbie > -----Original Message----- > From: CCP4 bulletin board <CCP4BB@JISCMAIL.AC.UK <mailto:CCP4BB@JISCMAIL.AC.UK> <mailto:CCP4BB@JISCMAIL.AC.UK> > On Behalf Of anna > anna > Sent: Tuesday, September 8, 2020 11:28 > To: CCP4BB@JISCMAIL.AC.UK <mailto:CCP4BB@JISCMAIL.AC.UK> <mailto:CCP4BB@JISCMAIL.AC.UK> > Subject: [ccp4bb] metal coordination at low resolution - restraints > > Dear all, > > quickly: is there a way to restrain metal coordination geometry (even angles) > in refmac? > > I am refining a low resolution structure (3.3A) with 2 zinc binding sites. > I am pretty sure about metal position (strong anomalous signal) and what > are the residues involved in coordination since I solved the apo- structure at > good resolution and Zn-binding does not induce huge structural variations. > However, as you can imagine, electron density is poorly defined and Refmac > gives a very distorted coordination geometry. > I noticed that in phenix it is possible to generate restraints with readyset but > I'd like to work with refmac. > > Many thanks for your suggestions. > > Cheers, > Anna > > ________________________________ > > > 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 <http://www.jiscmail.ac.uk/CCP4BB> <http://www.jiscmail.ac.uk/CCP4BB> , a mailing list hosted by www.jiscmail.ac.uk <http://www.jiscmail.ac.uk/> <http://www.jiscmail.ac.uk <http://www.jiscmail.ac.uk/>> , terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/------------------------------------------------------------------------ To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1-- Jan Dohnalek, Ph.DInstitute of Biotechnology Academy of Sciences of the Czech Republic Biocev Prumyslova 595 252 50 Vestec near Prague Czech Republic Tel. +420 325 873 758-- Jan Dohnalek, Ph.DInstitute of Biotechnology Academy of Sciences of the Czech Republic Biocev Prumyslova 595 252 50 Vestec near Prague Czech Republic Tel. +420 325 873 758------------------------------------------------------------------------ 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
######################################################################## 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/
