On 09/06/2008, Roberto Mosca <[EMAIL PROTECTED]> wrote: > From the document: > > "3.5 Resolution macromolecular sequence conflicts > Differences in entity sequence assignment between RCSB PDB and MSD-EBI > have > been resolved. Any remaining differences between the chemical sequence > and the > macromolecular sequence have also been resolved." > > I think that this refers only to the DBREF and SEQRES records in the > file. Of > course an alanine cannot become a valine or a tyrosine when going from > the original > to the remediated PDB.
I think it can. Lots of 'strange' format inconstancies have been cleaned up, including the unusual practice of labelling amino acids such as tyrosine as alanine if the majority of the side chain is 'unobserved' in the structure. In such cases the tyrosine may look like an alanine (chemically), and was sometimes labelled alanine in the ATOMSEQ. Tons of minor changes were made throughout the PDB files when going from PDB to PDB-REMEDIATED, however, these various changes were not documented... Amazing isn't it? Stupefying? I would be very happy if someone could prove me wrong on this point. It would be possible to get a database of differences by dumping the mmCIF files into XML and then doing an XML diff on all the pairs of files. The PDB's view seems to be 'just start using the remediated files', which you should do if possible. It is also surprising to learn that data in the PDB files such as links to sequence databases are not being kept up to date. For up to date sequence data you need to additionally use the 'sifts' (Structure integration with function, taxonomy and sequence) database: http://www.ebi.ac.uk/msd-srv/docs/sifts/ Hrm... I should add SIFTS to MetaBase ;-) Dan. > > I'm not sure what kind of errors have been fixed but I think there > could have been > inconsistencies between the sequence reported in the SEQRES records > and the > sequence corresponding to the ATOM records in the PDb file itself. > > Roberto > > ............................................ > Dr. Roberto Mosca > > EMBL c/o DESY > Notkestr. 85 > 22607 Hamburg > Germany > > Email: [EMAIL PROTECTED] > Tel: +49 (0)40 89902 131 > Web: http://www.embl-hamburg.de/~rmosca > ............................................ > > > > On Jun 8, 2008, at 04:03, Xue Li wrote: > >> Hello all, >> >>> From remediated pdb web site http://remediation.wwpdb.org/ >>> index.html, I read >> "Updated references to databases and taxonomies and Resolved >> differences >> between chemical and macromolecular sequences". >> >> Does it mean that for some proteins they will have difference >> sequences in >> remediated pdb and ordinary pdb? >> >> Thanks. >> >> -- >> Xue, Li >> Bioinformatics and Computational Biology program @ ISU >> Ames, IA 50010 >> 515-450-7183 >> _______________________________________________ >> BBB mailing list >> [email protected] >> http://www.bioinformatics.org/mailman/listinfo/bbb >> > > > _______________________________________________ > BBB mailing list > [email protected] > http://www.bioinformatics.org/mailman/listinfo/bbb > -- hello _______________________________________________ BBB mailing list [email protected] http://www.bioinformatics.org/mailman/listinfo/bbb
