Variability in automated assignment of NOESY spectra and three-dimensionalstructure determination: A test case on three small disulfide-bonded proteins

Three independent runs of automatic assignment and structure calculations w ere performed on three small proteins, calcicludine from the venom of the g reen mamba Dendroaspis angusticeps, kappa -conotoxin PVIIA from the purple cone Conus purpurascens and HsTX1, a short scorpion toxin from the venom of Heterometrus spinnifer. At the end of all the runs, the number of cross pe aks which remained unassigned (0.6%, 1.4% and 2% for calcicludine, kappa -c onotoxin and HsTX1, respectively), as well as the number of constraints whi ch were rejected as producing systematic violations (2.7%, 1.0%, and 1.4% f or calcicludine, kappa -conotoxin and HsTX1, respectively) were low. The co nformation of the initial model used in the procedure (linear model or cons tructed by homology) has no influence on the final structures. Mainly two p arameters control the procedure: the chemical shift tolerance and the cut-o ff distance. Independent runs of structure calculations, using the same par ameters, yield structures for which the rmsd between averaged structures an d the rmsd around each averaged structure were of the same order of magnitu de. A different cut-off distance and a different chemical shift tolerance y ield rmsd values on final average structures which did not differ more than 0.5 Angstrom compared to the rmsd obtained around the averaged structure f or each calculation. These results show that the procedure is robust when a pplied to such a small disulfide-bonded protein.