Geometry problem with sugars refined in phenix
Hi, I have refined a structure with NAGs linked to ASNs at 3 places in my structure. I used the right link definitions and the NAG definitions are from Phenix library. Phenix refined all the sugars right and made the right links between the NAGs. However, while submitting, PDB found problem with the geometry of NAGs linked to the ASN at two places. This is the issue -the atoms around C1 atom of the sugar molecule NAG A #810 and NAG A #1670 are on a plane and the C1 atom center is far off to be tetrahedral. Please, verify the chirality of C1 and provide the correct coordinates if necessary.- Any ideas on how to fix the geometry of these sugars will be of great help. Regularization and real space refine in Coot is not working or I don't know how to do it. I am looking for alternative ways to fix this problem. Thank You Tirumal
Tirumal Can you send me the PDB file, the link definitions you used and the .geo file from the final refinement. Please send it directly to me and it will be keep in he strictest confidence. Nigel On 3/18/10 9:56 AM, Tirumala Kumar Chowdary wrote:
Hi,
I have refined a structure with NAGs linked to ASNs at 3 places in my structure. I used the right link definitions and the NAG definitions are from Phenix library.
Phenix refined all the sugars right and made the right links between the NAGs. However, while submitting, PDB found problem with the geometry of NAGs linked to the ASN at two places.
This is the issue -the atoms around C1 atom of the sugar molecule NAG A #810 and NAG A #1670 are on a plane and the C1 atom center is far off to be tetrahedral. Please, verify the chirality of C1 and provide the correct coordinates if necessary.-
Any ideas on how to fix the geometry of these sugars will be of great help.
Regularization and real space refine in Coot is not working or I don't know how to do it. I am looking for alternative ways to fix this problem.
Thank You
Tirumal
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb
-- Nigel W. Moriarty Building 64R0246B, Physical Biosciences Division Lawrence Berkeley National Laboratory Berkeley, CA 94720-8235 Phone : 510-486-5709 Email : [email protected] Fax : 510-486-5909 Web : CCI.LBL.gov
I am working with sugar polymers and have run into several problems. First, is anyone working on improving the handling of sugars? ReadySet is getting most of the chiral restraints wrong for the sugars that don't already have parameters. It appears to convert all sugars to a standard chair conformation. For example, residue IDS gets 3 of 5 chiral centers inverted, even when starting from an ideal-geometry PDB file containing all hydrogens. It seems like a serious flaw for automated ligand parameterization to make such an obvious error. Another problem is that it is difficult to get the polymer links working, because many sugars don't have the same hydrogen nomenclature. Until that changes, we need better handling of the current residues. In most cases, only the hydrogen names are mismatched, so it would be great if PHENIX could ignore hydrogen atom errors in the case where hydrogens will not be used in refinement. Can I give PHENIX "data_link_list" CIF data just for some residues, or will it have to contain the entire existing list? Another problem is with LINK records. I tried to work around the polymer link problems by simply supplying LINK records. Unfortunately, LINK record processing requires REMARK 290 symmetry data, even if all entries have the identity symmop "1555". That is an unnecessary nuisance, especially since almost all LINK records have 1555. I discovered that leaving the symmop field blank works without REMARK 290, but why not make "1555" behave the same way. Additionally, how is the bond sigma derived? Does the link distance value have any effect? Thanks, Joe Krahn
participants (3)
-
Joe Krahn -
Nigel W Moriarty -
Tirumala Kumar Chowdary