Re: how to flag entries with serious problems

Paula Fitzgerald (paula_fitzgerald@Merck.Com)
Wed, 1 Nov 95 15:30:26 EST 

Frances Bernstein writes:

>      In the PDB format document there is a new record type
> specified that has not yet been implemented.  It would seem
> useful to have an equivalent in an mmCIF but I could not find
> any in the dictionary.  Here are the relevant parts of the PDB
> decsription:
> 
> CAVEAT warns of severe errors in an entry. Use caution when
> working with this entry.
> 
> COLUMNS        DATA TYPE       FIELD          DEFINITION
> 1 - 6          Record name     "CAVEAT"
> 9 - 10         Continuation    continuation   Allows concatenation of
>                                               multiple records
> 12 - 15        IDcode          idCode         PDB ID code of this entry
> 20 - 70        String          comment        Free text giving the
>                                               reason for the CAVEAT
> 
> PDB will add this record to incorrect entries that are not withdrawn
> from the set of released entries. This record will be used sparingly,
> and only after an external review has been made.
> 
> CAVEAT will be added by the PDB to entries known to be incorrect.
> 
> Example:
> 
>          1         2         3         4         5         6         7
> 1234567890123456789012345678901234567890123456789012345678901234567890
> CAVEAT     1ABC    THE CRYSTAL TRANSFORMATION IS IN ERROR BUT IS
> CAVEAT   2 1ABC    UNCORRECTABLE AT THIS TIME

Since this is a PDB specific issue, we decided to handle this by creating
a category, DATABASE_PDB_CAVEAT.  This is *almost* parallel to 
DATABASE_PDB_REMARK - the lack of parallelism is be illustrated by the
examples. 

For DATABASE_PDB_REMARK, we have:

;
     loop_
    _database_PDB_remark.id
    _database_PDB_remark.text
    3
    ;
    REMARK   3                                                              5HVP
    REMARK   3 REFINEMENT. BY THE RESTRAINED LEAST-SQUARES PROCEDURE OF J.  5HVP
    REMARK   3  KONNERT AND W. HENDRICKSON (PROGRAM *PROLSQ*).  THE R       5HVP
    ;
    4
    ;
    REMARK   4                                                              5HVP
    REMARK   4 THE TWO CHAINS OF THE DIMERIC ENZYME HAS BEEN ASSIGNED THE   5HVP
    REMARK   4 THE CHAIN INDICATORS *A* AND *B*.                            5HVP
    ;

while for DATABASE_PDB_CAVEAT, we have

     loop_
    _database_PDB_caveat.id
    _database_PDB_caveat.text
    1
    ;
    CAVEAT     1ABC    THE CRYSTAL TRANSFORMATION IS IN ERROR BUT IS
    ;
    2
    ;
    CAVEAT   2 1ABC    UNCORRECTABLE AT THIS TIME
    ;

the distinction being that for REMARK the id is extracted from the PDB record,
while for CAVEAT is is simply a running number made up to provide a label for
each line.  This could be changed if is was felt to be unncessarily clumsy -
the alternative would be a making CAVEAT not a looped list, but a single
data item into which all CAVEAT records were read.

Fran may want to comment on or rewrite the definition for the category, which
I have just made as boilerplate.

;              Data items in the DATABASE_PDB_CAVEAT category record details
               about features of the data block flagged as 'caveats' by the
               Protein Data Bank (PDB).

               These data items are included only for consistency with PDB
               format files.  They should appear in a data block only if that
               data block was created by reformatting a PDB format file.
;

Paula

********************************************************************************
 Dr. Paula M. D. Fitzgerald  ______________ voice and FAX: (908) 594-5510
   Merck Research Laboratories ______________ email: paula_fitzgerald@merck.com
     P.O. Box 2000, Ry50-105     ______________ or bean@merck.com           
       Rahway, NJ 07065  USA 
         (for express mail use 126 E. Lincoln Ave. instead of P. O. Box 2000)  
********************************************************************************