Try to detect TLS group exclusion of waters

[touwwouter]

http://proteincrystallography.org/ccp4bb/message15289.html
Message about REFMAC entries and TLS group definitions.
Summary:
Older (which versions?) REFMAC default: TLSD WATERS ADD
Result: waters only silently added to TLS groups during refinement but not included in header. After TLSANL: still residual B-factors for waters

Newer REFMAC default: TLSD WATERS EXCLUDE
Result: waters added to TLS groups during refinement and included in header. After TLSANL: full B-factors for waters

Detection possible?
e.g. 3npd

REMARK   3  OTHER REFINEMENT REMARKS: 1.HYDROGENS HAVE BEEN ADDED IN THE        
REMARK   3  RIDING POSITIONS. 2.A MET-INHIBITION PROTOCOL WAS USED FOR          
REMARK   3  SELENOMETHIONINE INCORPORATION DURING PROTEIN EXPRESSION. THE       
REMARK   3  OCCUPANCY OF THE SE ATOMS IN THE MSE RESIDUES WAS REDUCED TO 0.75   
REMARK   3  FOR THE REDUCED SCATTERING POWER DUE TO PARTIAL S-MET               
REMARK   3  INCORPORATION. 3.ATOM RECORDS CONTAIN SUM OF TLS AND RESIDUAL B     
REMARK   3  FACTORS. ANISOU RECORD CONTAINS SUM OF TLS AND RESIDUAL U           
REMARK   3  FACTORS. 4.WATERS WERE EXCLUDED FROM AUTOMATIC TLS ASSIGNMENT.      
REMARK   3  5.SULFATE (SO4) AND N-CYCLOHEXYL-3-AMINOPROPANESULFONIC ACID        
REMARK   3  (CAPS) FROM THE CRYSTALLIZATION SOLUTION AND 1,2-ETHANEDIOL (EDO)   
REMARK   3  FROM THE CRYOPROTECTANT SOLUTION HAVE BEEN MODELED IN THE SOLVENT   
REMARK   3  STRUCTURE.