Hi,
I have gone back and re-ran an old Refmac 5 job using CCP46.1.1 and
noticed dramatic differences on the final Rfree / Rwork and geometric
parameters. The job was re-ran using the "re-run" button on CCP4, same
input files (.mtz, .pdb) same number of cycles, etc.
The geometry restraints and weights for both jobs were identical and are
listed below (copied from the log file):
**** Geometry restraints and weights ****
Sigma:
Bonding distances
Weight = 0.50
Bond angles
Weight = 0.50
Planar groups
WEIGHT= 1.00
Chiral centers
Weight= 0.50
NON-BONDED CONTACTS
Overall weight = 1.00
Sigma for simple VDW = 0.20
Sigma for VDW trhough torsion angle = 0.20
Sigma for HBOND = 0.20
Sigma for metal-ion = 0.20
Sigma for DUMMY and other atom = 0.30
Distance for donor-accepetor = vdw1+vdw2+(-0.30)
Distance for acceptor - H = vdw1+ ( 0.10)
VDW distance through torsion = vdw1+vdw2+(-0.30)
Distance for DUMMY-others = vdw1+vdw2+(-0.70)
TORSION ANGLES
Weight= 1.00
THERMAL FACTORS
Weight= 1.00
Main chain bond (1-2 neighbour) 1.5A**2
Main chain angle (1-3 neighbour) 2.0A**2
Side chain bond 3.0A**2
Side chain angle 4.5A**2
RESTRAINTS AGAINST EXCESSIVE SHIFTS
Positional parameters 0.00A
Thermal parameters 0.00A
Occupancy parameters 0.00
RADIUS OF CONFIDENCE
Positional parameters 0.30A
Thermal parameters 0.03A**2
Occupancy parameters 0.50
The new Refmac5 is version 5.5.0072.
The old Refmac5 is version 5.2.0019.
The final line on the "new" refinement job is:
Ncyc Rfact Rfree FOM -LL -LLfree rmsBOND
zBOND rmsANGL zANGL rmsCHIRAL $$
30 0.1581 0.2037 0.874 245851. 8051.9 0.0468
1.958 3.088 1.500 0.296
The final line on the "old" refinement job is:
Ncyc Rfact Rfree FOM LLG rmsBOND rmsANGL
rmsCHIRAL $$
30 0.171 0.204 0.865 257924.2 0.025
2.088 0.155
Things very different on the "new" version:
- The rmsANGL shot up (from 2.088 to 3.088);
- The rmsBOND also shot up (from 0.25 to 0.0468).
As a consequence of having my geometric constrains blown up, the Rwork
went considerably down.
Does anybody have any thoughts on why using the same geometry restraints
led to very different results?
Cheers,
Rafael.
--
Rafael Couñago
Postdoctoral Research Fellow
Department of Biochemistry
University of Otago
710 Cumberland St
Dunedin, New Zealand
ph: (03) 479 5148
