3D structures for InSane may be read from any of the file formats supported by Wit!P. Proton names should follow the traditional X-Plor nomenclature, i.e. backbone amide protons are named HN, methylene protons are numbered 1,2 (e.g. HB1,HB2). Most often, the structure(s) will be available in PDB format, or as a combination of MOL2 and .cnf files. Independent of what file format is used, the first step of an InSane analysis in Wit!P is always the creation of an NMR data set:
Wit!P> nmr foobarwhere foobar is an arbitrary name. Next, the structures are loaded with the structure read command. E.g., to load a set of conformations stored in foobar_#.pdb (#=1,2,3,4,...):
NMR>
NMR> structure read pdb foobar_?.pdb foobar_??.pdbStructures are read into NMR data set in alphabetic order, followed by a "fit to average" superposition of the C-alpha positions of residues 87 to 122 (this step is optional: the InSane procedure is independent of the relative positions and orientations of the structure in the ensemble). The the end command terminates the structure input and returns to the NMR> prompt.
NMR structure option: fit average
molecules to fit to average: *
[0 atoms selected] template atoms: *_87-*122/ca
NMR structure option: end
NMR>
Alternatively, multiple conformations may be read from multiple model PDB files, of from .cnf files, e.g.:
NMR> structure read mol2 foobar.mol2(to use foobar.mol2 as the template for conformations from the file homology.cnf).
NMR structure option> confa homology.cnf *
NMR structure option: end
NMR>
Note: If the structures are read from PDB files with incomplete CONECT records, it may be necessary to add missing bonds manually:
NMR> bond auto *which is slightly dangerous, if the geometry of the structures is unusual (short non bonded contact, long bonds).