The information contained on the chain level includes the name and type of each chain in the structure, and the covalent links that connect it to other chains. All of this information is defined with the CHAIN statement. The covalent links may connect the chain to a symmetry image of another chain as well as to a direct image.
A CHAIN statement must be entered for each chain. The form of the CHAIN statement is:
CHAIN <Chain name> <Chain type> N(<Chain link>) <Chain link> :== <Residue name> <Chain name>|<Residue name> <Link type>
The first pair of tokens after the keyword gives the name of the chain and its type. The only information known about the chain type is its sequence. This information is presented on a series of RESIDUE statements, each of which mentions that chain type.
Each CHAIN statement allows the definition of links to other chains in the structure. The link is defined by first giving the name of the source residue. Then the chain and residue name of the target of the link is specified in the standard fashion by separating the two names with a ``|''. Finally, the link's type is given. This sequence of information may be repeated as many times as necessary to define all the links.
TNT will assume that each link connects the residue to the nearest symmetry image of the target. It is not required that the particular symmetry operator be specified.
If a chain has a link to a symmetry image of itself that link should be defined on the CHAIN statement as though the connection was to another chain. Bonds across symmetry axes cannot be defined on RESIDUE statements.