SEARCHDB -- Sequence search of the Brookhaven Protein Databank

I.  CREATING THE SEARCHDB DATABASE

         SEARCHDB achieves its speed by searching a condensed binary form
     of the Databank.  This must be created before SEARCHDB is run.  Note
     that this creation need be done once, and subsequently only when the
     Brookhaven Database is updated.

     To create this binary file (PDB.SEQUENCE):

     (1) Create a list of sequence files in PDB.LIST.  Can be done with

	 $ DIR/NOSIZE/NODATE/NOHEAD/NOTRAIL/OUT=PDBLIST.DAT DISK6:[PDB]PDB*.ENT;

         (Substitute your directory name for DISK6:[PDB])

     (2) Run MAKEDB to create PDB.SEQUENCE.  This takes about 2 1/2 minutes,
	 and should be a batch job.  Look at the log for a list of non-standard
	 residues found.  This is for informational purposes - the program deals
	 correctly with the sequences contining these.


II.  SEARCHING

     SETUPDB.COM defines symbols and logical names for the programs to run.
     It will need to be modified to refer to your directories.

     The search program is called SEARCHDB.  It runs interactively and will
     prompt for search sequences, or it can be run with a command line argument.
     Search sequences are single letter code.

     Wild Card Matches:
       X or * in a search sequence will match any amino acid.
       B and Z will match ASX and GLX respectively.


III.  BUGS AND MISCELLANEOUS

     (1) Residue numbers may be approximate if (a) sequence numbers do not
	 start with 1, or (b) sequence has insertions/deletions

     (2) To search some subset of the database, edit the PDB.LIST used
	 in creating the database and run MAKEDB on the edited file.
	 You can rename the resulting PDB.SEQUENCE, and just change
	 the logical name assignment when you want to search it.

     (3) With no modification to the program, you can search a subset
	 of the database (e.g. exclude unreleased coordinates).
	 Just edit the list of sequence files from step I.(1) above
	 before running MAKEDB to create an alternate SEARCHDB database.