I got rid of the protG example, primarily because it wasn't useful for showing people how to use marvin to solve structures.
The peak tables included in the CVN and IL-4 examples contain many bad peaks. They were generated by automatic peak picking using Dan Garrett's CAPP program. After the initial matching of peaks to shift assignments and filtering the resulting peak assignments, about 10-15% of the assigned peaks are completely wrong. More importantly, a very large number of wrong peak assignments are still attached to good peaks (i.e., peaks that have another, correct, peak assignment). And as we showed in our 2005 JACS paper, adding wrong peak assignments to otherwise-good peaks is, for the behavior of marvin, entirely equivalent to adding completely new, entirely wrong peaks. If you're really interested, the protG example is still available in older versions of xplor-nih. --JK On May 15, 2007, at 11:04 AM, Gary S. Thompson wrote: > Dear all > > in the eginput directory it states in the readme > > (9) marvin > contains an example of a method of weeding out incorrect NOEs. > This > example > uses protein G and randomly-assigned NOEs as well as the > experimental NOEs. > This example uses the TCL scripting language. > > On a recently downloaded xplor nih distribution (2.17.0) marvin/ > protg > is empty except for a CVS directory. > > however, marvin does contain cvn and il4 and I have managed to get > cvn/runMarvinQuick to run on our cluster and it did converge and > produce > a result. But, here is the question, does the cvn data dontain bogus > nOes and if not where can I get the protG test data so as to test this > feature? > > > regards > gary > > > -- > ------------------------------------------------------------------- > Dr Gary Thompson > Astbury Centre for Structural Molecular Biology, > University of Leeds, Astbury Building, > Leeds, LS2 9JT, West-Yorkshire, UK Tel. +44-113-3433024 > email: garyt at bmb.leeds.ac.uk Fax +44-113-2331407 > ------------------------------------------------------------------- > > > _______________________________________________ > Xplor-nih mailing list > Xplor-nih at nmr.cit.nih.gov > http://dcb.cit.nih.gov/mailman/listinfo/xplor-nih
