Guido Leoni wrote: > I'm dealing with a little big problem (for me). > I need to model a dymer protein. My structures are 952, 958, 898 > residues long and my sequence is formed by 990aa. > > When Modller run send my this error message: crsplin_223E> Internal > error: array too small: MAXSPL current maximum, current need: > 400 420. > Someone can help my?

I can't really answer your question properly without seeing your input files. See http://salilab.org/modeller/manual/node11.html

Usually, however, this means that Modeller has tried to convert a multi-Gaussian restraint to a cubic spline, but for some reason the original restraint spans such a large range that Modeller's internal limit (400) on the number of spline points is exceeded. You may be able to make the problem go away by turning off the spline conversion procedure entirely (by setting automodel.spline_on_site to False) or by increasing the spacing of the spline points from the default (0.5) by setting environ.spline_dx. However, the fact that your restraint spans such a large range (200A if your features are distances) is concerning, and this should probably be fixed instead.

Ben Webb, Modeller Caretaker