By default, proteins are re-sequenced into the standard PDB file sequence, i.e., all atoms in a residue are contiguous, when keywords ADD-H or SITE are used, or if the input data set is a pure PDB file. To prevent ADD-H or SITE from re-sequencing the data, add NORESEQ. To stop a pure PDB file from being resequenced, convert it to a MOPAC data set by adding the standard three lines to the start, and use keywords ADD-H and NORESEQ.
Some PDB files contain severe topologic errors that confuse the residue recognition routine, a pre-requisite step for resequencing. Such errors result in a warning message being printed, and the run stopped. This type of problem occurs most often when adding hydrogen atoms using ADD-H. By default ADD-H uses the residue recognition routine, and a topologic problem could prevent ADD-H from working correctly. To avoid this problem, run ADD-H with NORESEQ; this avoids using the residue recognition routine. Of course, the topologic problem is still present, even after hydrogen atoms are added. To correct the underlying problem, first optimize the positions of the hydrogen atoms, then optimize the positions of the atoms responsible for the topologic error.
Unless corrective action is taken, some topologic errors can result in an incorrect assignment of hydrogen atoms when using ADD-H. To correct this fault, use CVB with ADD-H and NORESEQ. This is an unusual use of CVB in that it normally cannot be used with ADD-H, but it is justified in this specific case.
See also: RESEQ, RESIDUES, XENO, CHAINS, and START_RES.