Dear amber users:
My system consists of a dimer-protein and one sugar in 10A water box, and the sugar dosent has any covalent bond with the protein. I met a problem of huge different DIHED when I tried to use MMPBSA to calculate the free energy though single or triple trajectories. DIHED of complex is 5xxx and the protein is 11xxx. (the other energy terms look fine)
So I went back to check my procedures of preparing the system and tried to find some information from the mail list and FAQ. Some people mantioned that the different scaling factor may cause the energy variation. But as David Case and Rose walker said, in Amber11 and Ambertool 1.5, sander would assign the scaling factor (scee=1.2,scnb=2.0) automatically.
In my output file, it showed the information below.
--------------------------------------------------------------------------------------
| Note: 1-4 EEL scale factors were NOT found in the topology file.
| Using default value of 1.2.
| Note: 1-4 VDW scale factors were NOT found in the topology file.
| Using default value of 2.0.
--------------------------------------------------------------------------------------
So I think the different DIHED may not be attributed to the scaling factor.
Another possibility of the different DIHED may be the overlap of the Glycam06 and ff99SB. I checked the parameter files of these two ff and I found that only one atom type "Cy" has different difinetions in these two ff. One is sp2 and the other one is sp3. But both my complex and protein dont have the atom type "Cy". So I think the overlapping atom type "Cy" may not casue the different DIHED neither.
When I checked my procedures of preparing the system and tried different source ff order to generate my top and rst file. I found something strange. If I source ff99SB first and then source Glycam_06, the DIHED of complex became 5xxx. But if I source Glycam_06 first and then ff99SB, it became 11xxx !!
Although my apo protein system comprises no glycan, I tried this in my apo protein too. If I source only ff99SB, the DIHED was 11xxx. If I source ff99SB first and then Glycam_06, the DIHED was 5xxx. But if I source Glycam_06 first and then ff99SB, it became 11xxx again.
In both apo and complex, the DIHED 11xxx looks more appropriate, because my dimer-protein comprises almost 1000 residues But I still wonder what causes this different DIHED and ensure there is nothing inappropriate for my later mmpbsa calculation.
Best regards :)
Ken
--------------------------------------------------------------------------------------
These are my operation steps in sleap
source leaprc.ff99SB
source leaprc.Glycam_06
com = loadpdb xxx.pdb
addions com 38 Na+ 12 Cl-
solvatebox com TIP3PBOX 10.0
saveamberparm com xxx.top xxx.rst
--------------------------------------------------------------------------------------
Here is my input file of min Complex.
minimise
&cntrl
imin=1,maxcyc=1000,ncyc=500,
cut=10.0,ntb=1,
ntpr=100,
ntr=1, restraintmask=':1-1114.CA,C,N',
restraint_wt=5.0
nmropt=1,
&end
&end
&wt type='END'
&end
DISANG=dist.rst
-------------------------------------------------------------------------------------
Output file 1 (source Glycam_06 and then ff99SB)
FINAL RESULTS
NSTEP ENERGY RMS GMAX NAME NUMBER
1000 -4.9221E+05 4.9197E-01 4.8894E+01 CE1 4127
BOND = 33940.8786 ANGLE = 2723.5512 DIHED = 4600.0170
VDWAALS = 70077.5842 EEL = -661208.6465 HBOND = 0.0000
1-4 VDW = 4062.8436 1-4 EEL = 52654.8763 RESTRAINT = 936.5455
EAMBER = -493148.8955
NMR restraints: Bond = 1.869 Angle = 0.000 Torsion = 0.000
----------------------------------------------------------------------------------------
Output file 2 (source ff99SB and then Glycam_06)
FINAL RESULTS
NSTEP ENERGY RMS GMAX NAME NUMBER
1000 -5.2555E+05 4.2406E-01 4.0278E+01 CD 14860
BOND = 37003.3845 ANGLE = 2663.2636 DIHED = 10799.4054
VDWAALS = 77597.0227 EEL = -711210.8832 HBOND = 0.0000
1-4 VDW = 4054.2862 1-4 EEL = 52593.7994 RESTRAINT = 947.6604
EAMBER = -526499.7214
NMR restraints: Bond = 1.686 Angle = 0.000 Torsion = 0.000
---------------------------------------------------------------------------------------
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sun Sep 18 2011 - 23:00:05 PDT
