-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Hello Krish-- > > Thanks for the fix on pdbLoad... > > I have been using the solution scattering potl. (following the examples on > the help pages). Without the solution scattering term all work well. > Whenever I use the solnScattering I get the following error: > "SystemError: xplor-nih error: only derivatives of uniform calcType are > supported". The protein is highly flexible and SAXS data has no significant > secondary structural details. (e.g part of the code at the end). > > If I am justified to use such a SAXS profile? If 'yes' is there a fix? > You may be justified in using the term, although an ensemble representation might be required. Anyway, the error message you get doesn't have anything to do with that aspect: For computational efficiency the SAXS data needs to be sampled uniformly in q. Prior to Xplor-NIH version 2.25, the datafile given to create_solnXRayPot had to have this uniform sampling. However, with the latest version, you can specify the numQ parameter to create_SolnXRayPot and it will use linear interpolation to get uniform sampling. The exact value to use for numQ varies, but if the SAXS curve is quite smooth, you might use something like numQ=15 or so. Then the second term created with create_SolnXRayPot might have numQ=30 for validation that the chi^2 or rmsd doesn't depend strongly on this number. If numQ=15 doesn't work, you can try increasing this value. best regards-- Charles -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Processed by Mailcrypt 3.5.8+ <http://mailcrypt.sourceforge.net/> iEYEARECAAYFAkuK3MQACgkQPK2zrJwS/lZ+5QCeNFq2nHrqxtOnULAXxGZdxQ5G G58AnizCzcvq1qNBQPYwT1Bk+vpwP+DG =v5Zc -----END PGP SIGNATURE-----
