Menu
▾
▴
[Psisoft-devl] Re: pdb-l: RE: mistakes in PDB files? (fwd)
|
From: Dan B. <dm...@mr...> - 2004-08-14 04:34:50
|
Regarding fixing problems with PDB files... (note that I reported a problem with pdb 1ona, not 1aon). ---------- Forwarded message ---------- Date: Fri, 13 Aug 2004 17:28:58 -0400 From: Christine Zardecki <zar...@rc...> To: Dan Bolser <dm...@mr...> Cc: "'in...@rc...'" <in...@rc...> Subject: Re: pdb-l: RE: mistakes in PDB files? Dear Dan, Thanks for your comments -- we will fix the error in 1aon with the next update of the PDB. PDB users wishing to contribute corrections to PDB entries should send this information to in...@rc.... We're working on a news item to describe this in greater detail, but some information is below. Sincerely, Christine Zardecki RCSB Protein Data Bank Corrections to entries orginally processed by the RCSB, EBI/MSD, or PDBj are handled by the PDB annotation staff and subsequently reviewed by the author(s) depositing the structure. Any changes in released PDB entries are described in the PDB REVDAT records and in the mmCIF/XML category DATABASE_PDB_REV_RECORD categories. In certain cases replacement coordinates for an entry are provided by the depositing author. In these cases the original entry is obsoleted and the replacement coordinates are released in a new superseding PDB entry. The relationship between obsolete and superseding entries is stored in OBSLTE/SPRSDE PDB records and in the mmCIF/XML category PDBX_DATABASE_PDB_OBS_SPR. Queries of obsoleted entries on the RCSB/PDB website always produce the most recent superseding entry. Obsoleted entries remain available in a separate area of the PDB ftp site, ftp://ftp.rcsb.org/pub/pdb/data/structures/obsolete/. For the entries deposited prior to 1998, a variety of consistency checks have been performed. This has been done as part of an ongoing project to maintain uniformity within the PDB archive. This effort is described in detail at http://www.rcsb.org/pdb/uniformity and in (ref data uniformity papers). Examples of uniformity corrections include corrections related to atomic nomenclature for both macromolecule and ligand, sequence-coordinate consistency, and the addition of missing records (e.g. citations, synonyms, and sequence database references). Corrections in the pre-1998 entries have been made only in the mmCIF and XML data files. The mmCIF and XML data files are download options of the RCSB PDB website and are also available via ftp from the following RCSB PDB servers: ftp://ftp.rcsb.org/pub/pdb/data/structures/all/mmCIF ftp://beta.rcsb.org/pub/pdb/uniformity/data/XML The XML data files were produced as part of a joint project by all wwPDB members, and these files are in the final stage of beta testing. Both mmCIF and XML data files conform to the PDB Exchange data dictionary. This dictionary is available in both mmCIF and XML schema form at http://deposit.pdb.org/mmcif/. On Aug 9, 2004, at 8:02 PM, Dan Bolser wrote: > On Fri, 6 Aug 2004, Oscar Hur wrote: > >> >> Hi >> >> Is it common to find mistakes in published PDB files? I just found >> quite >> a few mistakes in one of the proteins in this database. Is there >> anyway >> for Protein Data Bank to update/correct the existing file in its >> database? >> Is there a form for me to submit? What is the procedure to do so? > > It would be really nice to have a list of known problems and a > mechanism of reporting them. A simple 'bug tracker' would work for > these cases and would let people easily access a list of PDB files > with outstanding format issues. > > It is the little things which would be nice to 'que up' for someone to > check - and if necessary assign to 'not a bug' status. > > For example the PDB 1ona has a problematic assignment of hetero atoms > to chains, having a Ca and an Mn assigned to chain B which clearly > belong in chain C and another Ca and Mn assigned to chain C which > clarly belong in chain D and another Ca and Mn assigned to chain D > which clearly belong in chain B. The chain A Ca and Mn are correctly > assigned. > > > >> >> For example, in 1KNB, all the strands are anti-parallel. But in its >> PDB >> files: >> >> SHEET 1 V 6 THR 400 TRP 402 0 >> SHEET 2 V 6 ASP 418 LYS 427 1 >> SHEET 3 V 6 SER 430 ALA 440 1 >> SHEET 4 V 6 ASN 479 ASN 482 1 >> SHEET 5 V 6 LEU 485 THR 486 1 >> SHEET 6 V 6 TYR 573 TYR 577 1 >> SHEET 1 R 4 SER 454 PHE 461 0 >> SHEET 2 R 4 ASN 515 TYR 521 1 >> SHEET 3 R 4 LYS 528 THR 535 1 >> SHEET 4 R 4 TYR 550 TRP 556 1 >> >> The senses of the strands are all incorrectly labelled as parallel as >> indicated in col 39-40. They should be labelled as "-1". >> >> Oscar >> >> >> >> >> >> >> >> >> >> >> >> TO UNSUBSCRIBE OR CHANGE YOUR SUBSCRIPTION OPTIONS, please see >> https://lists.sdsc.edu/mailman/listinfo.cgi/pdb-l . >> > > > TO UNSUBSCRIBE OR CHANGE YOUR SUBSCRIPTION OPTIONS, please see > https://lists.sdsc.edu/mailman/listinfo.cgi/pdb-l . |
