Hi Paul, I just rebuilt coot-0.9.8.94, coot-0.9.8.93 and coot-0.9.8.92 from scratch, deleting the previous installs.
The glyco validator in version coot-0.9.8.94 is indeed fixed. Thank you so much! Also, coot-0.9.8.93 and coot-0.9.8.92 now build an alpha1-6 fucose correctly into the electron density (not sure why I got beta1-6 before), producing identical coordinates. However, building the same fucose in coot-0.9.8.94 fails. First, the carbohydrate moiety is inserted into a different orientation compared to coot-0.9.8.93. During the fit, it is being moved to the correct density blob, however it is detached completely from the carbohydrate tree. Distance between C1 (FUC) and O6 (NAG) results into 3.56 Å! After it is built, pressing "Refine tree" does not include the fucose and also exporting the tree does not contain it. I could trace the problem partially down to the library file share/coot/data/cho-acedrg/FUC-acedrg.cif If I remove this file from the installation so that coot-0.9.8.94 cannot load it, the fucose builds correctly, similar (but not identical) position to that from coot-0.9.8.93. I can get the same result (correct build) by importing share/coot/lib/data/monomers/f/FUC.cif *after* FUC-acedrg.cif had been loaded. I can re-induce broken behaviour (failed build) by re-importing FUC-acedrg.cif I believe the two libraries overwrite each other. Curiously enough, coot-0.9.8.93 contains the same libraries (identical versions), but I don't experience this behaviour, i.e. the fucose is built correctly no matter which library was loaded last. I have attached the carbohydrate trees, one built by coot-0.9.8.93, the other by coot-0.9.8.94 (detached fucose included). Btw. one more funny things I noticed was that the Refmac monomer libraries are loaded from $cleaned_dir: coot-0.9.8.94/coot-Linux-GenuineIntel---gtk2-python/share/coot/lib/data but the acedrg libraries are loaded from $fat_dir: coot-0.9.8.94/coot-gtk2-python/share/coot/data (not that it matters - except for debugging...) Thanks again! With best regards, Markus On 2024-04-23 20:23, Engin Özkan wrote:
Caution! This message was sent from outside the University of Manitoba. I see. Thanks, Paul, for the explanation. Of course you are right, and I see that too despite having the correct ALPHA1-6: find_glycosidic_linkage_type() for A 911 NAG,A 913 FUC returns "BETA1-6" This settles it for me. (I am not sure how Markus got beta-fucoses built by Coot the in the first place, though.) Engin On 4/23/24 8:03 PM, Paul Emsley wrote:We are talking about different things, I think. Coot's glyco builder builds alpha 1-6 linked FUC correctly. Coot's glyco validator used to (until this evening) identify the alpha 1-6 linked FUC as a beta anomer (it no longer does). See for example the G8 FUC in 8q5u: Coot 0.9.8.94 now prints: ASN [spec: 1 "A" 297 ""] pyr-ASN NAG [spec: 1 "G" 1 ""] BETA1-4 NAG [spec: 1 "G" 2 ""] BETA1-4 BMA [spec: 1 "G" 3 ""] ALPHA1-3 MAN [spec: 1 "G" 4 ""] BETA1-2 NAG [spec: 1 "G" 5 ""] BETA1-6 MAN [spec: 1 "G" 6 ""] BETA1-2 NAG [spec: 1 "G" 7 ""] ALPHA1-6 FUC [spec: 1 "G" 8 ""] Paul. On 24/04/2024 01:38, Engin Özkan wrote:CAUTION: This email originated from outside of the LMB: .-owner-c...@jiscmail.ac.uk-. Do not click links or open attachments unless you recognize the sender and know the content is safe. If you think this is a phishing email, please forward it to phish...@mrc-lmb.cam.ac.uk -- Hi Markus, I am using Coot that comes with CCP4 8, which has FUC (alpha-L-Fucose) with an oxygen on C1 in the axial position. Which is the correct form, I believe. Using Coot 0.9.8.93, I get my alpha1-6 and alpha1-3 (insect) Fucoses inserted correctly. I had trouble understanding your description: You mean you get Coot to insert a FUC with the glycosidic bond going equatorial, instead of axial, to Fucose? This is how it works for me: https://www.dropbox.com/scl/fi/yjbzstojjivmfpqdlz2os/Screen-Recording-2024-04-23-at-7.32.14-PM.mov?rlkey=zvvk7apd19wk61kt21qyc42l5&dl=0 I'll see what happens when I update to the new Coot version where Paul has switched the logic. By the way, privateer tells me I have zero errors and the correct alpha anomer for Fucose. Fuc913-a6-GlcNAc911-b-ASN258 WURCS=2.0/2,2,1/[a2122h-1b_1-5_2*NCC/3=O][a1221m-1a_1-5]/1-2/a6-b1 ... PDB Sugar Q Phi Theta Detected type Cnf <Bfac> Ctx Ok? ---- ------------ ----- ------ ------ ------------------- --- ------ --- ----- test NAG-A-911 0.570 280.48 5.29 beta-D-aldopyranose 4c1 47.88 (n) yes test FUC-A-913 0.535 92.53 177.32 alpha-L-aldopyranose 1c4 70.76 (n) yes ... Wrong anomer: 0 Best, Engin On 4/18/24 4:35 PM, Markus Meier wrote:Hi Engin, hmm, it appears I was using the monomer library from https://www2.mrc-lmb.cam.ac.uk/personal/pemsley/coot/dependencies/monomers-2023-01-02-23:57:29.tar.gz (it gets installed automatically with Coot's build-it script). However, the files in question (FUC.cif, NAG.cif, BMA.cif and MAN.cif) are identical to current CCP4 8.0.018 version of the library (installed by CCP4 Update Manager) as well as version ccp4-8.0.018 and version ccp4-8.0.016 on GitHub (https://github.com/MonomerLibrary/monomers) MD5 checksums on these files are 9cfba4ac24f056d648df8227316f788b monomers/f/FUC.cif 958ac5b19a92ceec604d233457944b2a monomers/n/NAG.cif 697dd5242334dd99dad9ce43133c6382 monomers/b/BMA.cif 9d68ff54296ef2f4622005efb4eb080b monomers/m/MAN.cif If they are different on your machine, please let me know. I will switch monomer libraries anyway and do some testing. I also was switching back and forth between my locally compiled coot (0.9.8.92) and the the CCP4 distributed version (0.9.8.93 EL). I will check if there is a difference between the two. Thank you very much for your suggestion! With best regards, Markus On 2024-04-18 13:48, Engin Özkan wrote:Caution! This message was sent from outside the University of Manitoba. Hi Markus, There were reports to the ccp4bb and phenixbb back in 2011 about something similar. The monomer library files for FUC and/or the linkage definitions for refmac/phenix were not right. I think. My recollection is that this was corrected, and I could not reproduce your observation. I also had private discussions with Garib Murshudov, who were correcting the glycan monomer problems back in ~2011. I could find old links to some of the discussions on bulletin board archives: https://phenix-online.org/pipermail/phenixbb/2011-November/017958.html https://www.mail-archive.com/ccp4bb@jiscmail.ac.uk/msg13968.html I am curious if something got reverted or if you have some ancient library definitions somewhere. But I hope this gives some ideas. Best, Engin On 4/17/24 3:07 PM, Markus Meier wrote:Hi Paul, thank you for implementing the Glyco module in Coot! I used its "Add N-linked Glycan" function to build an N-linked glycan which contains L-fucose with an alpha1-6 glycosidic linkage on the first N-acetyl glucosamine of the tree. However, after exporting the tree, I checked it with privateer and unfortunately privateer flags the L-fucose as having the wrong anomeric linkage - beta 1-6 :-( Coot reports the same as being in the alpha 1-6 anomeric form. I then manually built the opposite anomeric form which privateer happily accepts as being alpha1-6. However, Coot reports it as beta 1-6. In my understanding, the alpha configuration of a glycosidic linkage should have the oxygen at the anomeric centre and the -CH3 group at opposite faces of the sugar ring, so I have to agree with privateer on this one. Please check you code. I have attached the two glycan trees and here is the output of privateer and Coot for both. Coot and privateer versions are given at the bottom. Btw. Neither privateer nor refmac5 can read the mmCIF files that Coot produces with the "Extract this Tree" functions. All the _atom_site.label_*_id entries are empty! N133_tree_privateer_alpha_anomer_FUC.cif: Privateer: Detailed validation data ------------------------ PDB Sugar Q Phi Theta Detected type Cnf <Bfac> Ctx Ok? ---- ------------ ----- ------ ------ ------------------- --- ------ --- ----- N133 NAG-A-459 0.597 63.27 2.85 beta-D-aldopyranose 4c1 28.09 (n) yes N133 NAG-A-460 0.603 145.39 2.15 beta-D-aldopyranose 4c1 36.34 (n) yes N133 BMA-A-461 0.550 333.80 4.39 beta-D-aldopyranose 4c1 45.25 (n) yes N133 MAN-A-462 0.628 6.26 5.03 alpha-D-aldopyranose 4c1 46.22 (n) yes N133 MAN-A-463 0.628 4.87 5.87 alpha-D-aldopyranose 4c1 50.37 (n) yes N133 FUC-A-464 0.615 162.25 174.51 alpha-L-aldopyranose 1c4 31.24 (n) yes <- Partially occupied monosaccharides, if any, are marked with an asterisk (*) SUMMARY: Wrong anomer: 0 Wrong configuration: 0 Unphysical puckering amplitude: 0 In higher-energy conformations: 0 Privateer has identified 0 issues, with 0 of 6 sugars affected. Coot output: DEBUG:: find_glycosidic_linkage_type() number of sorted distances: 1 #### glyco close: 1.486 A 459 C1 to A 133 ND2 debug:: find_glycosidic_linkage_type() for A 459 NAG,A 133 ASN returns "pyr-ASN" DEBUG:: find_glycosidic_linkage_type() number of sorted distances: 4 #### glyco close: 2.397 A 459 C4 to A 460 C1 #### glyco close: 2.339 A 459 O4 to A 460 C2 #### glyco close: 2.318 A 459 O4 to A 460 O5 #### glyco close: 1.387 A 459 O4 to A 460 C1 debug:: find_glycosidic_linkage_type() for A 459 NAG,A 460 NAG returns "BETA1-4" DEBUG:: find_glycosidic_linkage_type() number of sorted distances: 3 #### glyco close: 2.347 A 460 O4 to A 461 C2 #### glyco close: 2.346 A 460 O4 to A 461 O5 #### glyco close: 1.398 A 460 O4 to A 461 C1 debug:: find_glycosidic_linkage_type() for A 460 NAG,A 461 BMA returns "BETA1-4" DEBUG:: find_glycosidic_linkage_type() number of sorted distances: 1 #### glyco close: 1.407 A 461 O6 to A 462 C1 debug:: find_glycosidic_linkage_type() for A 461 BMA,A 462 MAN returns "ALPHA1-6" DEBUG:: find_glycosidic_linkage_type() number of sorted distances: 1 #### glyco close: 1.403 A 461 O3 to A 463 C1 debug:: find_glycosidic_linkage_type() for A 461 BMA,A 463 MAN returns "ALPHA1-3" DEBUG:: find_glycosidic_linkage_type() number of sorted distances: 2 #### glyco close: 2.385 A 459 O6 to A 464 O5 #### glyco close: 1.399 A 459 O6 to A 464 C1 debug:: find_glycosidic_linkage_type() for A 459 NAG,A 464 FUC returns "BETA1-6" <- ########################################################################################## N133_tree_privateer_beta_anomer_FUC.cif: Privateer: Detailed validation data ------------------------ PDB Sugar Q Phi Theta Detected type Cnf <Bfac> Ctx Ok? ---- ------------ ----- ------ ------ ------------------- --- ------ --- ----- N133 NAG-A-459 0.539 93.90 14.28 beta-D-aldopyranose 4c1 17.31 (n) yes N133 NAG-A-460 0.576 3.57 3.48 beta-D-aldopyranose 4c1 28.29 (n) yes N133 BMA-A-461 0.534 312.39 9.79 beta-D-aldopyranose 4c1 42.24 (n) yes N133 MAN-A-462 0.580 219.67 4.19 alpha-D-aldopyranose 4c1 42.02 (n) yes N133 MAN-A-463 0.563 294.23 4.86 alpha-D-aldopyranose 4c1 47.39 (n) yes N133 FUC-A-464 0.542 120.60 172.19 beta-L-aldopyranose 1c4 18.87 (n) no <- Partially occupied monosaccharides, if any, are marked with an asterisk (*) SUMMARY: Wrong anomer: 1 Wrong configuration: 0 Unphysical puckering amplitude: 0 In higher-energy conformations: 0 Privateer has identified 1 issues, with 1 of 6 sugars affected. Coot: DEBUG:: find_glycosidic_linkage_type() number of sorted distances: 1 #### glyco close: 1.462 A 459 C1 to A 133 ND2 debug:: find_glycosidic_linkage_type() for A 459 NAG,A 133 ASN returns "pyr-ASN" DEBUG:: find_glycosidic_linkage_type() number of sorted distances: 4 #### glyco close: 2.355 A 459 C4 to A 460 C1 #### glyco close: 2.316 A 459 O4 to A 460 O5 #### glyco close: 2.313 A 459 O4 to A 460 C2 #### glyco close: 1.387 A 459 O4 to A 460 C1 debug:: find_glycosidic_linkage_type() for A 459 NAG,A 460 NAG returns "BETA1-4" DEBUG:: find_glycosidic_linkage_type() number of sorted distances: 3 #### glyco close: 2.37 A 460 O4 to A 461 C2 #### glyco close: 2.344 A 460 O4 to A 461 O5 #### glyco close: 1.401 A 460 O4 to A 461 C1 debug:: find_glycosidic_linkage_type() for A 460 NAG,A 461 BMA returns "BETA1-4" DEBUG:: find_glycosidic_linkage_type() number of sorted distances: 1 #### glyco close: 1.408 A 461 O6 to A 462 C1 debug:: find_glycosidic_linkage_type() for A 461 BMA,A 462 MAN returns "ALPHA1-6" DEBUG:: find_glycosidic_linkage_type() number of sorted distances: 3 #### glyco close: 2.39 A 461 C3 to A 463 C1 #### glyco close: 2.389 A 461 O3 to A 463 C2 #### glyco close: 1.398 A 461 O3 to A 463 C1 debug:: find_glycosidic_linkage_type() for A 461 BMA,A 463 MAN returns "ALPHA1-3" DEBUG:: find_glycosidic_linkage_type() number of sorted distances: 1 #### glyco close: 1.414 A 459 O6 to A 464 C1 debug:: find_glycosidic_linkage_type() for A 459 NAG,A 464 FUC returns "ALPHA1-6" <- Program versions: Coot 0.9.8.93 EL (ccp4) on Gentoo Linux 64 bit Privateer version MKIV : 06/02/2 CCP4 8.0.019 on Gentoo Linux 64 bit With best regards, Markus-- Engin Özkan, Ph.D. Associate Professor Dept of Biochemistry and Molecular Biology University of Chicago http://ozkan.uchicago.edu######################################################################## 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/-- Engin Özkan, Ph.D. Associate Professor Dept of Biochemistry and Molecular Biology University of Chicago Phone: (773) 834-5498 http://ozkan.uchicago.edu ######################################################################## 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/
-- Markus Meier, Ph.D. Research Associate University of Manitoba Department of Chemistry 144 Dysart Road Winnipeg, MB, R3T 2N2, Canada Phone: +1 204 474 7172 E-mail: markus.me...@umanitoba.ca ######################################################################## 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/
N133_tree-coot-0.9.8.93.cif
Description: CIF chemical test
N133_tree-coot-0.9.8.94.cif
Description: CIF chemical test