Hi, Jason, Kenneth,
Thanks for your answer! I have used ff99SB FF to generate a set of new
prmtop files, no error occurs now. However, I have compared this result
with what I get from radiopt=0, and GB model respectively, I'm confused
with the ENPOLAR term (the cavity solvation free energy term, and
EDISPER should be dispersion term), this term doesn't change when I
alter the radii. Whereas, the SASA is relative to the atom radii. The
result is as below:
radiopt0 radiopt1
GB VDWAALS -46.0124 -46.0124
EEL -43.371 -43.371
EGB 1.7614 61.7614
ESURF -6.5164 -6.5164
DELTA G gas -89.3834 -89.3834
DELTA G solv 55.2449 55.2449
DELTA TOTAL -34.1385 -34.1385
PB VDWAALS -46.0124 -46.0124
EEL -43.371 -43.371
EPB 65.4239 71.5709
ENPOLAR -35.6264 -35.6264
EDISPER 72.2592 68.1717
DELTA G gas -89.3834 -89.3834
DELTA G solv 102.0567 104.1162
DELTA TOTAL 12.6733 14.7328
I only change the radiopt parameter and keep other parameters the
same. It sames that INP=2, cavity_offset = -0.5692, cavity_surften =
0.0378, and radiopt=1 can't give a value which is close to, even better
than the GB result. In addition, I did a test, I set INP=1,
SURFTEN=0.005, SURFOFF=0.86, then the result is similar to GB
calculation in mm_pbsa.pl.
mm_pbsa.pl "INP1 SURFTEN=0.0378, SURFOFF=-0.5692" "INP1
SURFTEN=0.005, SURFOFF=0.86"
GB GBTOT -33.69 -33.69
PB PBTOT -63.28 -30.04
Is that because I should change some other parameter to make radiopt=1
work correctly?
I would be appreciate for the helps!
Sincerely yours,
Wei Liu，
On 2015-04-22 01:39, Jason Swails wrote:
> I second Kenneth's suggestion. The only way to get radiopt to work
> "correctly" is to modify the PBSA source code and add support for the
> atom
> types in the new force fields yourself. To my knowledge, the radiopt
> radii
> have not been kept up-to-date with the latest versions of the Amber
> force
> field, so your options are the following:
>
> 1. Modify the PBSA source code to recognize the new atom types in
> whichever
> force field you are using
> 2. Use ff99SB instead, since those are the atom types that (I believe)
> were
> originally supported by radiopt
> 3. Use radiopt=0 and simply take the solvation radii from the prmtop
>
> I would opt for (3), since those are pretty good in my experience.
>
> As for inp=2 vs. inp=1 -- which one is "better" is still an open
> question.
> It's been my experience that inp=1 is more reliable, although some
> people
> have reported better results with inp=2 in their studies.
>
> Note that if you are using the latest version of MMPBSA.py, the
> -make-mdins
> and -use-mdins route is unnecessary for controlling inp=2 and radiopt
> --
> those can be set directly in the MMPBSA.py input file.
>
> HTH,
> Jason
>
>
> On Tue, Apr 21, 2015 at 11:28 AM, wliu <wliu.itcs.ecnu.edu.cn> wrote:
>
>> Hello, everyone,
>>
>> I'm learning MMPBSA.py recently, for the sake of calculating the
>> binding free energy of protein with an organic molecule. Though
>> AmberTools13 Reference Manual, to make the calculation more accurate,
>> I
>> choose INP=2 (which is default, and I have use -make-mdins option to
>> check the temporary mdin files) to decompose non-polar solvation
>> energy
>> into cavity and dispersion terms. In the meanwhile, I need to set
>> radiopt=1. My input file is written as:
>>
>> Input file for running PB and GB
>> &general
>> startframe=0, endframe=600, interval=10,
>> verbose =2,
>> # entropy =1,
>> strip_mask=:WAT:Na+
>> /
>> &gb
>> igb=5, saltcon=0.100
>> /
>> &pb
>> radiopt=1,
>> istrng=0.100,
>> indi =1,
>> /
>>
>> and I get the error information as:
>>
>> CalcError: /gpfs01/home/ymei/software/amber14/bin/mmpbsa_py_energy
>> failed with prmtop com_nobox.prmtop!
>> PB Bomb in pb_aaradi(): No radius assigned for atom 7 CB
>> 2C
>>
>> Exiting. All files have been retained.
>>
>> In addition, I did a test, no matter the ligand is a protein or an
>> organic molecule, and both in MMPBSA.py and mm_pbsa.pl, this error
>> will
>> occur. Can anyone tell me what this error information stand for and
>> how
>> to use radiopt = 1 correctly?
>> I would be appreciate for the helps!
>>
>> Sincerely yours,
>> Wei Liu，
>>
>> wliu.itcs.ecnu.edu.cn
>> East China Normal University
>> ShangHai，china，200062
>>
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Apr 23 2015 - 06:00:03 PDT
