Hello Charles, I am now using the CNS dna-rna-allatom topology and parameter files for my xplor calculations, which works without any further changes.
Best regards, Maria On 05/12/2010 09:36 PM, Charles at Schwieters.org wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > > > Hello Maria-- > > >> I'm refining an RNA hairpin structure with xplor, using parameter file >> nucleic.par. When uploading my final structures to the RCSB validation >> server I notice that many angles and bond lengths do not cluster about >> the mean values given in Clowney et al., JACS 1996, 118, 509-518, but >> lie up to 2?/0.02A away from them. This is due to angles/bonds that are >> defined by the same term in nucleic.par but do differ a little in >> reality, e.g. Guanine C2-N3 should be 1.323A, Adenine C6-N1 should be >> 1.351A and they are both set to 1.339A in nucleic par. >> I do realize that all of these deviating values are well within the >> 5?/0.05A limits but am just wondering if it is ok for the structures to >> have all angles/bonds of one type shifted systematically by 2?/0.02A. I >> noticed that I do not have this issue with cns, which defines every one >> of these angles/bonds separately. >> > Three bonds in three bases C2-N3 in GUA, C6-N1 in ADE and N3-C4 in CYT > share the same value, and it seems that this bond length value dates > from 1984, or earlier. I would welcome someone updating these values, > but note that this will probably require that chemical types for those > atoms be changed. > > best regards-- > Charles > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.10 (GNU/Linux) > Comment: Processed by Mailcrypt 3.5.9<http://mailcrypt.sourceforge.net/> > > iEYEARECAAYFAkvrAygACgkQPK2zrJwS/lYu8ACZAemserB6IFVdJLm+FRJmKwZ0 > NZMAn37XWmot+FmMicqBn3ObJ/fFDi4v > =CD0S > -----END PGP SIGNATURE----- >
