Hi Josh,
Thanks for the suggestions.
While CPU time is not an issue, the suggestions won't really get at why there is an aberrant behavior in the convergence, nor why the numbers with and without salt concentration are near identical. The salt should affect the numbers, even at this low salt concentration - at least according to the other codes.
I did do a GB calculation, also a non-linear PB calculation, and used DELPHI stand-alone, but the calculations didn't help explain the numbers below.
Best,
Jesper
On Jun 17, 2013, at 1:14 AM, Josh Berryman <the.real.josh.berryman.gmail.com> wrote:
> My guess is that your code and params are fine, but that the underlying
> physics of a continuum approximation is giving some problems. MMPBSA with
> DNA is a tricky proposition due to the (somewhat) structured nature of the
> water and salt shells around the molecule: at least, that is my own
> experience from a quick study on the B/Z transition, which admittedly
> happens at rather higher ionic strengths than you are looking at (M, not
> mM). If netropsin itself is multiply charged, then that won't be helping.
> A few more second opinions you can get are:
>
> 1) Do a GBSA calculation: the approximation to the PB equation which GB
> carries out isn't actually too bad, and this is easy to do.
> 2) Try using a fresh download of delPhi with default params instead of
> whatever is bundled with AMBER.
> 3) Try calling it through NAB instead of mmpbsa.py
> 4) Try some kind of TI approach, this will be expensive but I guess at UCSD
> you have plenty of CPU time, right?
>
> Josh
>
>
> On 17 June 2013 21:21, <deeptinayar.gmail.com> wrote:
>
>>
>> Sent from my BlackBerry® smartphone from !DEA
>>
>> -----Original Message-----
>> From: "Sorensen, Jesper" <jesorensen.ucsd.edu>
>> Date: Mon, 17 Jun 2013 05:33:48
>> To: AMBER Mailing List<amber.ambermd.org>
>> Reply-To: AMBER Mailing List <amber.ambermd.org>
>> Subject: [AMBER] MMPBSA results for DNA-ligand system
>>
>> Hi Amber experts,
>>
>> I have calculated the PB energies for the binding process between DNA and
>> Netropsin. However, I am getting some curious results that I'd like to
>> understand more and I'd like to hear if any of you have ideas?
>> I calculated these using PBSA stand-alone (see numbers below). Radii were
>> MBONDI, the remaining details are in the PBSA input options below.
>>
>> 1) The results below are calculated with and without salt concentration.
>> However, the results are very close to each other. I have compared the
>> results with other codes (e.g. APBS) and they get values diverging 8-10
>> kcal/mol due to the salt effect.
>>
>> 2) I have calculated the results using different grid resolutions
>> (focusing was turned off), yet the results don't really display what I
>> expected in terms of convergence behavior, but instead just fluctuate
>> around the same range of values. Is this the expected behavior for PBSA?
>>
>> Any insight would be really helpful.
>>
>> Best regards,
>> Jesper
>>
>>
>> Salt concentration = 100mM (istrng=100.000000)
>> GridRes dEPB
>> .15 1202.3177
>> .20 1201.9713
>> .25 1201.9845
>> .30 1202.5554
>> .35 1202.2611
>> .40 1202.0522
>> .45 1202.2218
>> .50 1201.9514
>> .55 1202.4602
>> .60 1201.7396
>> .65 1200.5449
>> .70 1200.9878
>> .75 1201.1727
>> .80 1200.8361
>> .85 1199.6109
>> .90 1201.1486
>> .95 1203.6329
>> 1.0 1202.2278
>>
>> Salt concentration = 0 mM
>> GridRes dEPB
>> .15 1201.9931
>> .20 1200.3285
>> .25 1200.1105
>> .30 1200.4338
>> .35 1200.0356
>> .40 1200.5173
>> .45 1201.1034
>> .50 1201.0788
>> .55 1201.0346
>> .60 1200.3855
>> .65 1199.6695
>> .70 1200.2409
>> .75 1201.1009
>> .80 1200.4463
>> .85 1199.1350
>> .90 1201.1621
>> .95 1203.6926
>> 1.0 1202.1964
>>
>>
>> PBSA input file:
>>
>> PBSA input
>> &cntrl
>> imin=0,
>> ntx=1,
>> ipb=1,
>> inp=1,
>> /
>> &pb
>> epsin=1.000000,
>> epsout=80.000000,
>> smoothopt=0,
>> istrng=100.000000,
>> pbtemp=300,
>> radiopt=0,
>> dprob=1.4,
>> iprob=2.00,
>> arcres=0.0625,
>> npopt=0,
>> solvopt=1,
>> accept=0.001,
>> maxitn=10000,
>> fillratio=2,
>> space=1.0,
>> nbuffer=0,
>> nfocus=1,
>> fscale=8,
>> npbgrid=1,
>> bcopt=5,
>> eneopt=2,
>> frcopt=0,
>> scalec=0,
>> sasopt=0,
>> saopt=0,
>> cutfd=5,
>> cutnb=0,
>> nsnba=1,
>> phiout=0,
>> phiform=1,
>> npbverb=1,
>> sprob=1.4,
>> rhow_effect=1.000,
>> use_sav=0,
>> cavity_surften=0.005420
>> cavity_offset=-1.008000,
>> maxsph=400
>> _______________________________________________
>> 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
>>
> _______________________________________________
> 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 Mon Jun 17 2013 - 08:00:02 PDT
