Simulations of proteins are best tackled using the package DLPROTEIN [46] which is an adaptation of DL_POLY specific to protein modelling. However you may simulate proteins and other macromolecules with DL_POLY_2 if you wish. This is described below.
If you select a protein structure from a SEQNET file (e.g. from the Brookhaven database), use the utility PROSEQ to generate the file CONFIG. This will then function as input for DL_POLY_2 . Some caution is required here however, as the protein structure may not be fully determined and atoms may be missing from the CONFIG file.
If you have the ``edit.out'' file produced by AMBER for your molecule use this as the CONNECT_DAT input file for the utility AMBFORCE. AMBFORCE will produce the DL_POLY_2 FIELD and CONFIG files for your molecule.
If you do not have the ``edit.out'' file things are a little more tricky, particularly in coming up with appropriate partial charges for atomic sites. However there are a series of utilities that will at least produce the CONNECT_DAT file for use with AMBFORCE. We now outline these utilities and the order in which they should be used.
If you have a structure from the Cambridge Structural database (CSDB) then use the utility FRACCON to take fractional coordinate data and produce a CONNECT_DAT and ``ambforce.dat'' file for use with AMBFORCE. Note that you will need to modify FRACCON to get the AMBER names correct for sites in your molecule. The version of FRACCON supplied with DL_POLY_2 is specific to the valinomycin molecule.
If you require an all atom force field and the database file does not contain hydrogen positions then use the utility FRACFILL in place of FRACCON. FRACCON produces an output file HFILL which should then be used as input for the utility HFILL. The HFILL utility fills out the structure with the missing hydrogens. (Note that you may need to know what the atomic charges are in some systems, for example the AMBER charges from the literature.)
Note: with minor modifications the utilities FRACFILL and FRACCON can be used on structures from databases other than the Cambridge structural database.