A simple DNA duplex

Here the input model is a DNA structure obtained from the PDB. The nomenclature used for residue names and some atom names is significantly different from that used in CNS. There is a utility program called fix_dna_rna which can be used to convert a PDB format DNA/RNA file to something more suitable for CNS:

      fix_dna_rna < 126D.pdb > 126D_fix.pdb [1 second]

The fixed coordinates of the model must be converted to a format suitable for use in CNS and a file describing the molecular topology of the model (MTF file) must also be generated. This is done with the CNS task file generate_easy.inp or for more complex cases with generate.inp.

      cns_solve < generate_easy.inp > generate_easy.out [7 seconds]

The ribose sugars in the CNS nucleic acid topology files are in the oxy form (ie. RNA). At the generate stage these riboses must be converted to DNA (deoxy form) if this is a DNA molecule. There is an atom selection which is used for this purpose:

{========================= RNA to DNA conversion  ==========================}

{* All nucleic acid residues initially have ribose sugars (rather than
   deoxyribose). A patch must be applied to convert the ribose to deoxyribose
   for DNA residues. Select those residues which need to have the patch
   applied to make them DNA. *}
{* Make sure that the atom selection is specific for the nucleic acid
   residues *}
{===>} dna_sele=(segid A or segid B);

There are two files generated: a PDB file and an MTF file (this contains the molecular topology information which describes to covalent topology of the molecule).