You can subscribe to this list here.
2002 
_{Jan}

_{Feb}

_{Mar}

_{Apr}

_{May}

_{Jun}

_{Jul}

_{Aug}
(1) 
_{Sep}
(2) 
_{Oct}

_{Nov}

_{Dec}
(6) 

2003 
_{Jan}
(11) 
_{Feb}
(5) 
_{Mar}
(4) 
_{Apr}
(9) 
_{May}
(14) 
_{Jun}
(10) 
_{Jul}
(2) 
_{Aug}
(6) 
_{Sep}

_{Oct}
(2) 
_{Nov}
(3) 
_{Dec}
(7) 
2004 
_{Jan}
(32) 
_{Feb}
(15) 
_{Mar}
(19) 
_{Apr}
(9) 
_{May}
(9) 
_{Jun}
(7) 
_{Jul}
(21) 
_{Aug}
(20) 
_{Sep}
(14) 
_{Oct}
(55) 
_{Nov}
(22) 
_{Dec}
(20) 
2005 
_{Jan}
(14) 
_{Feb}
(17) 
_{Mar}
(10) 
_{Apr}
(25) 
_{May}
(11) 
_{Jun}
(33) 
_{Jul}
(19) 
_{Aug}
(13) 
_{Sep}
(29) 
_{Oct}
(16) 
_{Nov}
(35) 
_{Dec}
(18) 
2006 
_{Jan}
(27) 
_{Feb}
(10) 
_{Mar}
(46) 
_{Apr}
(43) 
_{May}
(15) 
_{Jun}
(41) 
_{Jul}
(20) 
_{Aug}
(10) 
_{Sep}
(30) 
_{Oct}
(16) 
_{Nov}
(31) 
_{Dec}
(22) 
2007 
_{Jan}
(16) 
_{Feb}
(28) 
_{Mar}
(36) 
_{Apr}
(28) 
_{May}
(32) 
_{Jun}
(22) 
_{Jul}
(18) 
_{Aug}
(21) 
_{Sep}
(32) 
_{Oct}
(62) 
_{Nov}
(13) 
_{Dec}
(52) 
2008 
_{Jan}
(29) 
_{Feb}
(16) 
_{Mar}
(3) 
_{Apr}
(16) 
_{May}
(29) 
_{Jun}
(26) 
_{Jul}
(10) 
_{Aug}
(30) 
_{Sep}
(51) 
_{Oct}
(21) 
_{Nov}
(18) 
_{Dec}
(59) 
2009 
_{Jan}
(29) 
_{Feb}
(7) 
_{Mar}
(20) 
_{Apr}
(7) 
_{May}
(37) 
_{Jun}
(35) 
_{Jul}
(72) 
_{Aug}
(63) 
_{Sep}
(26) 
_{Oct}
(12) 
_{Nov}
(22) 
_{Dec}
(29) 
2010 
_{Jan}
(26) 
_{Feb}
(11) 
_{Mar}
(50) 
_{Apr}
(11) 
_{May}
(16) 
_{Jun}
(21) 
_{Jul}
(29) 
_{Aug}
(43) 
_{Sep}
(36) 
_{Oct}
(10) 
_{Nov}
(24) 
_{Dec}
(10) 
2011 
_{Jan}
(12) 
_{Feb}
(31) 
_{Mar}
(26) 
_{Apr}
(6) 
_{May}
(19) 
_{Jun}
(54) 
_{Jul}
(32) 
_{Aug}
(17) 
_{Sep}
(18) 
_{Oct}
(25) 
_{Nov}
(11) 
_{Dec}
(17) 
2012 
_{Jan}
(9) 
_{Feb}
(17) 
_{Mar}
(25) 
_{Apr}
(22) 
_{May}
(39) 
_{Jun}
(54) 
_{Jul}
(23) 
_{Aug}
(12) 
_{Sep}
(33) 
_{Oct}
(31) 
_{Nov}
(25) 
_{Dec}
(27) 
2013 
_{Jan}
(25) 
_{Feb}
(37) 
_{Mar}
(1) 
_{Apr}
(40) 
_{May}
(38) 
_{Jun}
(23) 
_{Jul}
(8) 
_{Aug}
(19) 
_{Sep}
(11) 
_{Oct}
(16) 
_{Nov}
(18) 
_{Dec}
(9) 
2014 
_{Jan}
(5) 
_{Feb}
(24) 
_{Mar}
(8) 
_{Apr}

_{May}
(44) 
_{Jun}
(14) 
_{Jul}
(15) 
_{Aug}
(16) 
_{Sep}

_{Oct}
(6) 
_{Nov}
(12) 
_{Dec}
(13) 
2015 
_{Jan}
(6) 
_{Feb}
(7) 
_{Mar}
(9) 
_{Apr}
(2) 
_{May}
(8) 
_{Jun}
(4) 
_{Jul}
(6) 
_{Aug}
(7) 
_{Sep}
(1) 
_{Oct}
(2) 
_{Nov}
(4) 
_{Dec}
(4) 
2016 
_{Jan}
(3) 
_{Feb}
(1) 
_{Mar}
(1) 
_{Apr}

_{May}

_{Jun}

_{Jul}

_{Aug}

_{Sep}

_{Oct}

_{Nov}

_{Dec}

2017 
_{Jan}

_{Feb}

_{Mar}

_{Apr}

_{May}
(1) 
_{Jun}
(4) 
_{Jul}
(1) 
_{Aug}

_{Sep}

_{Oct}

_{Nov}

_{Dec}

S  M  T  W  T  F  S 



1
(4) 
2

3
(1) 
4
(3) 
5

6

7
(1) 
8
(1) 
9
(1) 
10
(2) 
11
(5) 
12

13

14
(1) 
15
(4) 
16
(3) 
17

18

19
(1) 
20

21
(2) 
22

23

24

25

26

27

28

29

30

31



From: Nathan Baker <baker@bi...>  20091221 16:56:48

Hello  This looks like a good addition to the grid checking. Yong/Dave, can you test this out and see if it works with other features in the code? Thank you very much for your fix and help with APBS!  Nathan On Mon, Dec 21, 2009 at 8:58 AM, matteo.rotter <matteo.rotter@...>wrote: > Dear all, > > We are using apbs/1.2.1) for calculate the elec and apolar forces > acting on a molecule in solvent through the Poisson Boltzmann equation. > We use the mgmanual option to calulate first the results for a coarse > grid and then the results for a fine grid, smaller than the coarse one. > Center of the fine grid is different form the center of the coarse grid. > > What we find is that the fine grid didn't work due to the error: > Vpmg_dbForce: Atom 7 off grid! > VASSERT: ASSERTION FAILURE! filename routines.c, line 1205, > (Vpmg_dbForce(pmg, (*atomForce)[j].dbForce, j, pbeparm>srfm)) > Increasing glen little for the fine grid would bring to an error for > another atom, (now atom 7 is in the grid) but in my mind that wasn't a > good way to solve the problem.. > Looking in the apbs1.2.1source/src/mg/vpmg.c we suppose that the > problem is that the atom involved is in the grid, but his radius for > calculation goes out of it. > So we tried to find a way to solve the problem: > > Vpmg_dbForce: > line 5542: original: > /* Make sure we're on the grid */ > if ((apos[0]<=xmin)  (apos[0]>=xmax)  \ > (apos[1]<=ymin)  (apos[1]>=ymax)  \ > (apos[2]<=zmin)  (apos[2]>=zmax)) { > > > seems that would be better to calculate position < (xmin + rtot) and > > (xmax  rtot) to include also the dimensions of splineWin in the > calculation. > > /* Make sure we're on the grid */ > if ((apos[0]<=xmin + rtot)  (apos[0]>=xmax  rtot)  \ > (apos[1]<=ymin + rtot)  (apos[1]>=ymax  rtot)  \ > (apos[2]<=zmin + rtot)  (apos[2]>=zmax  rtot)) { > > > > Would also be better to include the radius of solvent in the calculation > of rtot: > rtot = (arad + thee>splineWin + srad); > > Now the calulation works without errors. > What do you think about our changes? > > Regards, > > Matteo. > > >  > This SF.Net email is sponsored by the Verizon Developer Community > Take advantage of Verizon's bestinclass app development support > A streamlined, 14 day to market process makes app distribution fast and > easy > Join now and get one step closer to millions of Verizon customers > http://p.sf.net/sfu/verizondev2dev > _______________________________________________ > apbsusers mailing list > apbsusers@... > https://lists.sourceforge.net/lists/listinfo/apbsusers >  Nathan Baker (http://bakergroup.wustl.edu) Associate Professor, Dept. of Biochemistry and Molecular Biophysics Director, Computational and Molecular Biophysics Graduate Program Center for Computational Biology, Washington University in St. Louis 
From: matteo.rotter <matteo.rotter@un...>  20091221 15:21:49

Dear all, We are using apbs/1.2.1) for calculate the elec and apolar forces acting on a molecule in solvent through the Poisson Boltzmann equation. We use the mgmanual option to calulate first the results for a coarse grid and then the results for a fine grid, smaller than the coarse one. Center of the fine grid is different form the center of the coarse grid. What we find is that the fine grid didn't work due to the error: Vpmg_dbForce: Atom 7 off grid! VASSERT: ASSERTION FAILURE! filename routines.c, line 1205, (Vpmg_dbForce(pmg, (*atomForce)[j].dbForce, j, pbeparm>srfm)) Increasing glen little for the fine grid would bring to an error for another atom, (now atom 7 is in the grid) but in my mind that wasn't a good way to solve the problem.. Looking in the apbs1.2.1source/src/mg/vpmg.c we suppose that the problem is that the atom involved is in the grid, but his radius for calculation goes out of it. So we tried to find a way to solve the problem: Vpmg_dbForce: line 5542: original: /* Make sure we're on the grid */ if ((apos[0]<=xmin)  (apos[0]>=xmax)  \ (apos[1]<=ymin)  (apos[1]>=ymax)  \ (apos[2]<=zmin)  (apos[2]>=zmax)) { seems that would be better to calculate position < (xmin + rtot) and > (xmax  rtot) to include also the dimensions of splineWin in the calculation. /* Make sure we're on the grid */ if ((apos[0]<=xmin + rtot)  (apos[0]>=xmax  rtot)  \ (apos[1]<=ymin + rtot)  (apos[1]>=ymax  rtot)  \ (apos[2]<=zmin + rtot)  (apos[2]>=zmax  rtot)) { Would also be better to include the radius of solvent in the calculation of rtot: rtot = (arad + thee>splineWin + srad); Now the calulation works without errors. What do you think about our changes? Regards, Matteo. 
From: Nathan Baker <baker@bi...>  20091219 00:38:11

Hello  This problem is due to an ambiguity in the user manual that has been corrected. Previously, the "usemap diel" command was documented ( http://apbs.wustl.edu/MediaWiki/index.php/ELEC_input_file_section#usemap) as Dielectric function map (as read by read diel); this causes the pdie, sdie, srad, swin, and srfm parameters and the radii of the biomolecular atoms to be ignored when computing dielectric values for the PoissonBoltzmann equation. This statement has been clarified to read Dielectric function map (as read by read diel); this causes the pdie, sdie, srad, swin, and srfm parameters and the radii of the biomolecular atoms to be ignored when computing dielectric maps for the PoissonBoltzmann equation. Note that the pdie and sdie values are still used for some boundary condition calculations as specified by bcfl. We have confirmed that the problem you describe is due to the effect of the "sdie" parameter on the boundary conditions. In other words, the code is behaving as expected but the documentation was confusing. Thank you for reporting this issue, Nathan 2009/12/16 <przemekbartha@...> > Hello, > I am doing same research on solvation energy of an elipsoid that has got a > dielectric constant equal 1, where surrounding area's constant is 80. > To do that, I implement my own "dx" maps. I have adapded one of examples to > my needs. > My quastion is: where does the difference between calculation #1 (solvated > state) and calculation #2 (reference state) come from, since I use the same > maps in both calculations? > The difference is about 57kJ/mol. The energies are respectively ~2323kJ/mol > and 2380kJ/mol. > Below, I attach my files. > > best regards, > Przemek > > > *APBS input file: * > read > mol pqr ion.pqr > diel dx 1.5_1.5_xelips_65.dx 1.5_1.5_yelips_65.dx 1.5_1.5_zelips_65.dx > # diel dx mapax.dx mapay.dx mapaz.dx > end > > elec name solv # Electrostatics calculation on the solvated > state > > > mgmanual # Specify the mode for APBS to run > usemap diel 1 > dime 65 65 65 # The grid dimensions > nlev 4 # Multigrid level parameter > grid 0.33 0.33 0.33 # Grid spacing > gcent mol 1 # Center the grid on molecule 1 > mol 1 # Perform the calculation on molecule 1 > lpbe # Solve the linearized PoissonBoltzmann > # equation > bcfl mdh # Use all multipole moments when calculating the > # potential > pdie 1.0 # Solute dielectric > sdie 80 # Solvent dielectric > chgm spl2 # Splinebased discretization of the delta > # functions > srfm mol # Molecular surface definition > > srad 1.4 # Solvent probe radius (for molecular surface) > swin 0.3 # Solvent surface spline window (not used here) > sdens 10.0 # Sphere density of accessibility object > temp 298.15 # Temperature > gamma 0.105 # Apolar energy surface coefficient (not used > here) > calcenergy total # Calculate energies > calcforce no # Do not calculate forces > write pot dx potentialRADIUS > # Write out the potential > # write dielx dx mapax > # write diely dx mapay > # write dielz dx mapaz > > end > > elec name ref # Calculate potential for reference (vacuum) > state > > > mgmanual > usemap diel 1 > dime 65 65 65 > nlev 1 > grid 0.33 0.33 0.33 > gcent mol 1 > mol 1 > lpbe > bcfl mdh > pdie 1.0 # The solvent and solute dielectric constants are > # equal > sdie 1.0 # The solvent and solute dielectric constants are > # equal > chgm spl2 > srfm mol > srad 1.4 > swin 0.3 > sdens 10.0 > temp 298.15 > gamma 0.105 > calcenergy total > calcforce no > > # write diely dx mapay > # write dielz dx mapaz > # write kappa dx kappa > > end > > print > > energy solv  ref > > end > > quit > > *ion.pqr* > ATOM 1 I ION 1 4.000 4.000 4.000 1.00 1.5 > > > > >  > This SF.Net email is sponsored by the Verizon Developer Community > Take advantage of Verizon's bestinclass app development support > A streamlined, 14 day to market process makes app distribution fast and > easy > Join now and get one step closer to millions of Verizon customers > http://p.sf.net/sfu/verizondev2dev > _______________________________________________ > apbsusers mailing list > apbsusers@... > https://lists.sourceforge.net/lists/listinfo/apbsusers > >  Nathan Baker (http://bakergroup.wustl.edu) Associate Professor, Dept. of Biochemistry and Molecular Biophysics Director, Computational and Molecular Biophysics Graduate Program Center for Computational Biology, Washington University in St. Louis 
From: Nathan Baker <baker@bi...>  20091216 17:42:40

Hi All  I agree with George's surmise: the nlev parameter can affect the precision of the final answers and is likely the culprit in this situation. Thanks, Nathan On Wed, Dec 16, 2009 at 11:06 AM, Gernot Kieseritzky < gernotf@...> wrote: > On Wed, 20091216 at 17:15 +0100, przemekbartha@... wrote: > > My quastion is: where does the difference between calculation #1 > > (solvated state) and calculation #2 (reference state) come from, since > > I use the same maps in both calculations? > > The difference is about 57kJ/mol. The energies are respectively > > ~2323kJ/mol and 2380kJ/mol. > > Below, I attach my files. > > >From your scripts: > > > elec name solv > > ... > > mgmanual # Specify the mode for APBS to run > > usemap diel 1 > > dime 65 65 65 # The grid dimensions > > nlev 4 # Multigrid level parameter > > versus > > > elec name ref > > > > > > mgmanual > > usemap diel 1 > > dime 65 65 65 > > nlev 1 > > The nlev is different in both calculations. In principle nlev=1 should > be compatible with dime = 65^3 but it could affect the numerics (but i > don't know much about the details of Multigrid algorithms). So try to > set them to an identical value. > > Best regards, > Gernot > > > >  > This SF.Net email is sponsored by the Verizon Developer Community > Take advantage of Verizon's bestinclass app development support > A streamlined, 14 day to market process makes app distribution fast and > easy > Join now and get one step closer to millions of Verizon customers > http://p.sf.net/sfu/verizondev2dev > _______________________________________________ > apbsusers mailing list > apbsusers@... > https://lists.sourceforge.net/lists/listinfo/apbsusers >  Nathan Baker (http://bakergroup.wustl.edu) Associate Professor, Dept. of Biochemistry and Molecular Biophysics Director, Computational and Molecular Biophysics Graduate Program Center for Computational Biology, Washington University in St. Louis 
From: Gernot Kieseritzky <gernotf@ch...>  20091216 17:06:55

On Wed, 20091216 at 17:15 +0100, przemekbartha@... wrote: > My quastion is: where does the difference between calculation #1 > (solvated state) and calculation #2 (reference state) come from, since > I use the same maps in both calculations? > The difference is about 57kJ/mol. The energies are respectively > ~2323kJ/mol and 2380kJ/mol. > Below, I attach my files. >From your scripts: > elec name solv > ... > mgmanual # Specify the mode for APBS to run > usemap diel 1 > dime 65 65 65 # The grid dimensions > nlev 4 # Multigrid level parameter versus > elec name ref > > > mgmanual > usemap diel 1 > dime 65 65 65 > nlev 1 The nlev is different in both calculations. In principle nlev=1 should be compatible with dime = 65^3 but it could affect the numerics (but i don't know much about the details of Multigrid algorithms). So try to set them to an identical value. Best regards, Gernot 
From: <przemekbartha@gm...>  20091216 16:15:28

Hello, I am doing same research on solvation energy of an elipsoid that has got a dielectric constant equal 1, where surrounding area's constant is 80. To do that, I implement my own "dx" maps. I have adapded one of examples to my needs. My quastion is: where does the difference between calculation #1 (solvated state) and calculation #2 (reference state) come from, since I use the same maps in both calculations? The difference is about 57kJ/mol. The energies are respectively ~2323kJ/mol and 2380kJ/mol. Below, I attach my files. best regards, Przemek APBS input file: read mol pqr ion.pqr diel dx 1.5_1.5_xelips_65.dx 1.5_1.5_yelips_65.dx 1.5_1.5_zelips_65.dx # diel dx mapax.dx mapay.dx mapaz.dx end elec name solv # Electrostatics calculation on the solvated state mgmanual # Specify the mode for APBS to run usemap diel 1 dime 65 65 65 # The grid dimensions nlev 4 # Multigrid level parameter grid 0.33 0.33 0.33 # Grid spacing gcent mol 1 # Center the grid on molecule 1 mol 1 # Perform the calculation on molecule 1 lpbe # Solve the linearized PoissonBoltzmann # equation bcfl mdh # Use all multipole moments when calculating the # potential pdie 1.0 # Solute dielectric sdie 80 # Solvent dielectric chgm spl2 # Splinebased discretization of the delta # functions srfm mol # Molecular surface definition srad 1.4 # Solvent probe radius (for molecular surface) swin 0.3 # Solvent surface spline window (not used here) sdens 10.0 # Sphere density of accessibility object temp 298.15 # Temperature gamma 0.105 # Apolar energy surface coefficient (not used here) calcenergy total # Calculate energies calcforce no # Do not calculate forces write pot dx potentialRADIUS # Write out the potential # write dielx dx mapax # write diely dx mapay # write dielz dx mapaz end elec name ref # Calculate potential for reference (vacuum) state mgmanual usemap diel 1 dime 65 65 65 nlev 1 grid 0.33 0.33 0.33 gcent mol 1 mol 1 lpbe bcfl mdh pdie 1.0 # The solvent and solute dielectric constants are # equal sdie 1.0 # The solvent and solute dielectric constants are # equal chgm spl2 srfm mol srad 1.4 swin 0.3 sdens 10.0 temp 298.15 gamma 0.105 calcenergy total calcforce no # write diely dx mapay # write dielz dx mapaz # write kappa dx kappa end print energy solv  ref end quit ion.pqr ATOM 1 I ION 1 4.000 4.000 4.000 1.00 1.5 
From: J.Dziedzic <J.Dziedzic@so...>  20091215 17:43:21

Gernot Kieseritzky wrote: > Hi! > [...] > In your file 'test.pqr' you have assigned a charge +6 corresponding to > the oxygen nucleus, and +1 corresponding to the protons. So the result > somehow suggests that such an arrangement of naked nuclei would be > stable in bulk water due to solvation. I guess, that's nonsense, > because forces other than electrostatics would prevent this. In reality, > of course, you have electrons around the nuclei  so you would have to > add the electrostatic energy due to the electrons to get the full > picture. Yes, of course. The arrangement I've suggested represents the naked cores of the oxygen and hydrogen atoms, as seen from the point of view of a DFTwithpseudopotential approach. The electrons are treated separately, they are represented by a continuous distribution of charge density, that integrates to a charge of 8, discretized on a grid. This part of my APBS calculation is progressing without doubts, that's why I have omitted it. thank you,  Jacek Dziedzic 
From: Gernot Kieseritzky <gernotf@ch...>  20091215 17:37:30

Hi! On Tue, 20091215 at 13:58 +0000, J.Dziedzic wrote: > >> \epsilon \Delta{}G E in dielectric of \epsilon > >> 80 27119 8785 > >> 50 26905 8571 > >> 30 26526 8192 > >> 10 24650 6316 > >> 5 21865 3531 > >> 4 20480 2146 > >> 3 18181 153 > >> 2 13606 4728 > >> 1 0 18334 > >> > >> What is the reason for my getting negative energies? Could that > >> be that the dielectric is polarizing sooo much? > > > > Well, it would suggest that the solvation energy in absolute values is > > larger than the Coulomb repulsion of your system. Sounds kind of counter > > intuitive with point charges of +6 and +1 at close distance. But I have > > the feeling you mixed up sdie and pdie (see above). > > I mixed them up only when typing the email, the input file was fine. I repeated your calculation at sdie = 80 and I think the values are fine (whether they are physically reasonable is another story). The very low solvation energy of 27119 is due to the high charge density creating a strong reaction field. Setting the partial charges to that of TIP3 water from the CHARMM force field roughly reproduces its solvation energy of about 10 kcal/mol. So there's no grid artefact at work here. In your file 'test.pqr' you have assigned a charge +6 corresponding to the oxygen nucleus, and +1 corresponding to the protons. So the result somehow suggests that such an arrangement of naked nuclei would be stable in bulk water due to solvation. I guess, that's nonsense, because forces other than electrostatics would prevent this. In reality, of course, you have electrons around the nuclei  so you would have to add the electrostatic energy due to the electrons to get the full picture. Best regards, Gernot Kieseritzky 
From: Gernot Kieseritzky <gernotf@ch...>  20091215 13:59:20

Hi! On Mon, 20091214 at 10:35 +0000, J.Dziedzic wrote: > Below I'm listing the \Delta{}G values produced by your script > along with the value of the total electrostatic energy in sdie=1, > pdie=\epsilon, obtained after \Delta{}G to the E1 value calculated above. Shouldn't it be sdie=\epsilon and pdie=1? Otherwise \Delta{}G would be the solvation energy of set of point charges in a small \epsilon=80 bubble surrounded by vacuum. > \epsilon \Delta{}G E in dielectric of \epsilon > 80 27119 8785 > 50 26905 8571 > 30 26526 8192 > 10 24650 6316 > 5 21865 3531 > 4 20480 2146 > 3 18181 153 > 2 13606 4728 > 1 0 18334 > > What is the reason for my getting negative energies? Could that > be that the dielectric is polarizing sooo much? Well, it would suggest that the solvation energy in absolute values is larger than the Coulomb repulsion of your system. Sounds kind of counter intuitive with point charges of +6 and +1 at close distance. But I have the feeling you mixed up sdie and pdie (see above). > On a side note  I thought of a simple approach to remove the grid > artefact, can you comment on the feasibility? > > 1) Output the charge distribution to an OpenDX file. > 2) Locate all 3x3x3 pockets of isolated charge floating in a sea of > zero charge. > 3) Calculate the Coulomb sum for the interaction of these 27 charges. > Since they will always be in a cavity, epsilon will be that of pdiel > and constant. Repeat for all 3x3x3 charge pockets. > 4) Subtract the obtained result from what APBS produces. > > This assumes that all point charges are separated by at least one > grid point of zero charge and that chgm spl2 is used. > > Is this feasible? Maybe. But in the end it's easier (and faster?) to simply subtract the homogenious APBS result. If electrostatics is really timelimiting and the grid artefact creates a serious problem in your application you should consider using a Generalized Born approach. The currently fastest implemention comes from the group of Prof. Caflisch: http://www3.interscience.wiley.com/journal/116327343/abstract?CRETRY=1&SRETRY=0 and is part of the latest CHARMM version 35. Best regards, Gernot Kieseritzky 
From: J.Dziedzic <J.Dziedzic@so...>  20091215 13:59:01

Gernot Kieseritzky wrote: > Hi! > > On Mon, 20091214 at 10:35 +0000, J.Dziedzic wrote: >> Below I'm listing the \Delta{}G values produced by your script >> along with the value of the total electrostatic energy in sdie=1, >> pdie=\epsilon, obtained after \Delta{}G to the E1 value calculated above. > > Shouldn't it be sdie=\epsilon and pdie=1? Otherwise \Delta{}G would be > the solvation energy of set of point charges in a small \epsilon=80 > bubble surrounded by vacuum. Sorry, a typo on my part. I got it right in the input file, so the calculation was for pdie=1.0 and varying sdie, as it should be. >> \epsilon \Delta{}G E in dielectric of \epsilon >> 80 27119 8785 >> 50 26905 8571 >> 30 26526 8192 >> 10 24650 6316 >> 5 21865 3531 >> 4 20480 2146 >> 3 18181 153 >> 2 13606 4728 >> 1 0 18334 >> >> What is the reason for my getting negative energies? Could that >> be that the dielectric is polarizing sooo much? > > Well, it would suggest that the solvation energy in absolute values is > larger than the Coulomb repulsion of your system. Sounds kind of counter > intuitive with point charges of +6 and +1 at close distance. But I have > the feeling you mixed up sdie and pdie (see above). I mixed them up only when typing the email, the input file was fine. >> On a side note  I thought of a simple approach to remove the grid >> artefact, can you comment on the feasibility? >> >> 1) Output the charge distribution to an OpenDX file. >> 2) Locate all 3x3x3 pockets of isolated charge floating in a sea of >> zero charge. >> 3) Calculate the Coulomb sum for the interaction of these 27 charges. >> Since they will always be in a cavity, epsilon will be that of pdiel >> and constant. Repeat for all 3x3x3 charge pockets. >> 4) Subtract the obtained result from what APBS produces. >> >> This assumes that all point charges are separated by at least one >> grid point of zero charge and that chgm spl2 is used. >> >> Is this feasible? > > Maybe. I have tried this approach. It doesn't really work, because there is a shift in the potential by a constant, I believe. APBS uses the qphi integration and my approach uses a sum over pairs of q's. When phi is shifted, the results do not match. Is there any way to determine the constant term in the potential? > But in the end it's easier (and faster?) to simply subtract the > homogenious APBS result. I agree. I would have used the approach you have suggested, but I was worried about the negative results I got for the three point charges. So, do you think this is a valid result, just counterintuitive? I am also worried that with this technique the result will depend on the precise dimensions of the cavity. > If electrostatics is really timelimiting and > the grid artefact creates a serious problem in your application you > should consider using a Generalized Born approach. The currently fastest > implemention comes from the group of Prof. Caflisch: > > http://www3.interscience.wiley.com/journal/116327343/abstract?CRETRY=1&SRETRY=0 > > and is part of the latest CHARMM version 35. Thank you. The electrostatic calculation is not a time bottleneck in my case, far from it. It's just that I need to be able to get reliable values for electrostatic energy for a set of point charges embedded in an inhomogenous dielectric. I have another doubt as well. When a point charge is spilled onto the grid, the grid artefact occurs because all of these grid points are now in the potential generated by their own collection of charges. This artefact is meant to cancel out when calculating \Delta{}G, because a similar scenario occurs in the inhomogeneous dielectric case. However, as the boundary conditions are the same for both cases, the addition of the dielectric around the cavity now implies a subtle change in the potential in the cavity to accomodate the fact, that the same boundary conditions must be satisfied. Would this not change the grid artefact energy subtly, leading to the cancelingout being not 100% thanks,  Jacek Dziedzic 
From: J.Dziedzic <J.Dziedzic@so...>  20091214 10:36:00

Thank you very much, Gernot Kieseritzky for a detailed explanation of how to subtract the grid artefact out. This method works fine for the trivial dimer case I have presented earlier, yet I'm seeing weird results for the next case on my list  a collection of three point charges that are the atomic cores of H2O. Of course I have no idea what the solvation energy is for such a system of isolated cores, yet I am worried that your approach seems to claim its total electrostatic energy with sdie=80, pdie=1 is negative. The input file is copied from your example: read mol pqr test.pqr end elec name test_inhom mgauto dime 65 65 65 nlev 4 cglen 50 50 50 fglen 12 12 12 cgcent mol 1 fgcent mol 1 mol 1 lpbe bcfl mdh pdie 1.0 sdie 80.0 chgm spl2 srfm smol swin 0.3 srad 1.4 sdens 10.0 temp 298.15 calcenergy total calcforce no end elec name test_hom mgauto dime 65 65 65 nlev 4 cglen 50 50 50 fglen 12 12 12 cgcent mol 1 fgcent mol 1 mol 1 lpbe bcfl mdh pdie 1.0 sdie 1.0 chgm spl2 srfm smol swin 0.3 srad 1.4 sdens 10.0 temp 298.15 calcenergy total calcforce no end print elecEnergy test_inhom  test_hom end quit ... with test.pqr now being REMARK 1 Cores of H2O ATOM 1 O XXX 1 12.000 12.957 12.000 6.00 1.50 ATOM 2 H XXX 1 12.000 12.000 12.000 1.00 1.20 ATOM 3 H XXX 1 12.927 13.197 12.000 1.00 1.20 The Coulomb energy with sdie=pdie=1 is E1=18334 kJ/mol as calculated by the coulomb.c utility. Calculation by hand gives a similar result of 18326 kJ/mol. Below I'm listing the \Delta{}G values produced by your script along with the value of the total electrostatic energy in sdie=1, pdie=\epsilon, obtained after \Delta{}G to the E1 value calculated above. \epsilon \Delta{}G E in dielectric of \epsilon 80 27119 8785 50 26905 8571 30 26526 8192 10 24650 6316 5 21865 3531 4 20480 2146 3 18181 153 2 13606 4728 1 0 18334 What is the reason for my getting negative energies? Could that be that the dielectric is polarizing sooo much? On a side note  I thought of a simple approach to remove the grid artefact, can you comment on the feasibility? 1) Output the charge distribution to an OpenDX file. 2) Locate all 3x3x3 pockets of isolated charge floating in a sea of zero charge. 3) Calculate the Coulomb sum for the interaction of these 27 charges. Since they will always be in a cavity, epsilon will be that of pdiel and constant. Repeat for all 3x3x3 charge pockets. 4) Subtract the obtained result from what APBS produces. This assumes that all point charges are separated by at least one grid point of zero charge and that chgm spl2 is used. Is this feasible? thank you,  J. 
From: Nathan Baker <baker@bi...>  20091211 23:24:19

Hello  APBS 0.3.2 is over 5 years old and there have been many changes since then. I would encourage you to review the revision history at http://www.poissonboltzmann.org/apbs/releasehistory; there are any number of reasons why the potential might be different. Sorry, Nathan On Dec 11, 2009, at 4:25 PM, johanna wrote: > PyMOLers, > > While switching computers our lab recently noticed major differences between electrostatic potentials calculated by an old version of apbs (version 0.3.2) and a new version (apbs 1.2) . What we see is the appearance of "charged holes" in the new version. > > What I mean by "charged holes" is that the inside of small cavities on the protein surface acquired strong surfaces charges somewhere between apbs version 0.3.2 and 1.2. > I have attached a png file of 3IKO chain C to illustrate the issue. All calculations were done at the default settings, via plugin in pymol and the program terminated without error. > > (428 KB) 3IKO_chainC.png > via the following link(s): > > http://elf.rockefeller.edu/get.php?f=8dee885168b6a6ba898d9b793c8d63af > > Expiration: Files will be removed on Dec 21, 2009 at 5:17 pm >  > > We have now a heated discussion about these "charged holes" in the lab and are wondering if anybody else noticed them, > > best regards, > Johanna > > > Johanna napetschnig > Graduate Fellow > The Rockefeller University > Blobel Lab / HHMI > >  > Return on Information: > Google Enterprise Search pays you back > Get the facts. > http://p.sf.net/sfu/googledev2dev > _______________________________________________ > apbsusers mailing list > apbsusers@... > https://lists.sourceforge.net/lists/listinfo/apbsusers  Nathan Baker (http://bakergroup.wustl.edu/) Associate Professor, Dept. of Biochemistry and Molecular Biophysics Director, Computational and Molecular Biophysics Graduate Program Center for Computational Biology, Washington University in St. Louis 
From: johanna <jnapetschn@ma...>  20091211 22:55:31

PyMOLers, While switching computers our lab recently noticed major differences between electrostatic potentials calculated by an old version of apbs (version 0.3.2) and a new version (apbs 1.2) . What we see is the appearance of "charged holes" in the new version. What I mean by "charged holes" is that the inside of small cavities on the protein surface acquired strong surfaces charges somewhere between apbs version 0.3.2 and 1.2. I have attached a png file of 3IKO chain C to illustrate the issue. All calculations were done at the default settings, via plugin in pymol and the program terminated without error. (428 KB) 3IKO_chainC.png via the following link(s): http://elf.rockefeller.edu/get.php?f=8dee885168b6a6ba898d9b793c8d63af Expiration: Files will be removed on Dec 21, 2009 at 5:17 pm  We have now a heated discussion about these "charged holes" in the lab and are wondering if anybody else noticed them, best regards, Johanna Johanna napetschnig Graduate Fellow The Rockefeller University Blobel Lab / HHMI 
From: Gernot Kieseritzky <gernotf@ch...>  20091211 17:43:27

Hi! Based on Dziedzic's example: read mol pqr test.pqr end elec name test_inhom mgauto dime 65 65 65 nlev 4 cglen 50 50 50 fglen 12 12 12 cgcent mol 1 fgcent mol 1 mol 1 lpbe bcfl mdh pdie 1.0 sdie 80.0 chgm spl2 srfm smol swin 0.3 srad 1.4 sdens 10.0 temp 298.15 calcenergy total calcforce no end elec name test_hom mgauto dime 65 65 65 nlev 4 cglen 50 50 50 fglen 12 12 12 cgcent mol 1 fgcent mol 1 mol 1 lpbe bcfl mdh pdie 1.0 sdie 1.0 chgm spl2 srfm smol swin 0.3 srad 1.4 sdens 10.0 temp 298.15 calcenergy total calcforce no end print energy test_inhom  test_hom end quit where test.pqr is ATOM 1 H XXX A 1 0.000 0.000 0.000 1.000 1.000 ATOM 2 H XXX A 1 0.529 0.000 0.000 1.000 1.000 Result (APBS 1.1): test_inhom  test_hom = 2497.9 kJ/mol. For the total electrostatic energy we've got to add the Coulomb energy with sdie=pdie=1. In his example we have: Ec = 1 Hartree = 2625.5 kJ/mol. (You can also use the "coulomb.c" program in the APBS tools directory of the source tar ball). So we obtain for the total electrostatic energy with sdie=80 and pdie=1: E = 2625.5  2497.9 kJ/mol = +127.6 kJ/mol. When sdie approaches 1, E should tend to the Coulomb's law result 2625.5 +/ numerical error (in the order of 1E13), because test_inhom  test_hom vanishes in the limit. Best regards, Gernot On Fri, 20091211 at 10:23 0500, Gatti, Domenico wrote: > Hi All, > Could we post an EXAMPLE SCRIPT of how the corrections suggested by > Gernot and Nathan for Dziedzic's simple calculation of two point charges > should be implemented? Does this mean that the grid artifact subtraction > should be carried out always in all the APBS calculations? Perhaps, this > occurs already by default, and I did not realize it. > Best, > Domenico > > > Domenico Gatti > Biochemistry & Mol. Biology > Wayne State University School of Medicine > 540 E. Canfield Avenue > Detroit, MI 48201 > Tel: 3135770620 or 3139934238 > Fax: 3135772765 > dgatti@... > > > > > > > Hi All  > > Gernot is absolutely correct. I would also add that, after correcting the > issues Gernot raised below, you should also examine the sensitivity of your > results on "chgm spl0" vs. "chgm spl2" since charge discretization can > affect these types of calculations as well. > > Thanks, > > Nathan > > On Dec 10, 2009, at 9:31 AM, Gernot Kieseritzky wrote: > > > Hi! > > > > On Wed, 20091209 at 17:49 +0000, J.Dziedzic wrote: > >> Hi! > >> > >> I am confused about the result of a very simple calculation when done > >> with APBS. Consider a trivial system of two point charges with unit > >> charges at a separation of 1 Bohr length, in vacuum. The Coulombic > >> energy of this system is exactly 1 Hartree, that is 2625.5 kJ/mole. > >> ... > >> ... APBS yields 31740 kJ/mole, which off by a factor of 12. Making the > >> grid finer only makes things worse, the results being: > >> > >> dime Energy (kJ/mole) > >> 65 1.209E04 > >> 129 2.222E04 > >> 193 3.174E04 > >> 257 4.058E04 > >> 289 4.539E04 > >> > >> which are all way off, by a factor of 520, from the correct value of > >> 2625.5 kJ/mole. > >> > >> I understand that normally one is interested in energy differences > >> between a system in vacuo and a solvated system, and any discretization > >> errors introduced are canceled if the grid is the same in both > >> calculations. Yet, with a system so trivial, without any dielectric > >> at all and, thus, without the arbitrariness of the cavity, what is the > >> underlying reason for the calculation being so off from the mark? > > > > Two words: grid artefact! Basically, the deviation is not due to > > numerical problems, rather the high energy values you observe are the > > result of the selfinteraction of the grid points. That's why the > > deviation is increasing with higher resolution as the grid points are > > getting closer. What you have to do to get the total electrostatic > > energy of your system without selfenergies: > > > > 1) Compute the Coulomb energy in the homogeneous continuum. > > > > 2) Compute the solvation energy of the same charge distribution using > > APBS. The grid artefact cancels as you calculate an energy difference > > here. This, of course, requires that you use the same grid setup in both > > APBS runs. > > > > 3) Add the values together. > > > > Best regards, > > Gernot Kieseritzky > > > >  > Return on Information: > Google Enterprise Search pays you back > Get the facts. > http://p.sf.net/sfu/googledev2dev > _______________________________________________ > apbsusers mailing list > apbsusers@... > https://lists.sourceforge.net/lists/listinfo/apbsusers 
From: Nathan Baker <baker@bi...>  20091211 17:22:31

Hello  Actually, all of the input files provided in the APBS examples/ directory demonstrate this in a variety of ways. Some of the examples remove selfenergies by calculating solvation energies while other examples remove selfenergies by using exactly the same grid positions for all atoms in each part of the calculation. Hope this helps, Nathan On Fri, Dec 11, 2009 at 9:23 AM, Gatti, Domenico <dgatti@...>wrote: > Hi All, > Could we post an EXAMPLE SCRIPT of how the corrections suggested by > Gernot and Nathan for Dziedzic's simple calculation of two point charges > should be implemented? Does this mean that the grid artifact subtraction > should be carried out always in all the APBS calculations? Perhaps, this > occurs already by default, and I did not realize it. > Best, > Domenico > > > Domenico Gatti > Biochemistry & Mol. Biology > Wayne State University School of Medicine > 540 E. Canfield Avenue > Detroit, MI 48201 > Tel: 3135770620 or 3139934238 > Fax: 3135772765 > dgatti@... > > > > > > > Hi All  > > Gernot is absolutely correct. I would also add that, after correcting the > issues Gernot raised below, you should also examine the sensitivity of your > results on "chgm spl0" vs. "chgm spl2" since charge discretization can > affect these types of calculations as well. > > Thanks, > > Nathan > > On Dec 10, 2009, at 9:31 AM, Gernot Kieseritzky wrote: > > > Hi! > > > > On Wed, 20091209 at 17:49 +0000, J.Dziedzic wrote: > >> Hi! > >> > >> I am confused about the result of a very simple calculation when done > >> with APBS. Consider a trivial system of two point charges with unit > >> charges at a separation of 1 Bohr length, in vacuum. The Coulombic > >> energy of this system is exactly 1 Hartree, that is 2625.5 kJ/mole. > >> ... > >> ... APBS yields 31740 kJ/mole, which off by a factor of 12. Making the > >> grid finer only makes things worse, the results being: > >> > >> dime Energy (kJ/mole) > >> 65 1.209E04 > >> 129 2.222E04 > >> 193 3.174E04 > >> 257 4.058E04 > >> 289 4.539E04 > >> > >> which are all way off, by a factor of 520, from the correct value of > >> 2625.5 kJ/mole. > >> > >> I understand that normally one is interested in energy differences > >> between a system in vacuo and a solvated system, and any discretization > >> errors introduced are canceled if the grid is the same in both > >> calculations. Yet, with a system so trivial, without any dielectric > >> at all and, thus, without the arbitrariness of the cavity, what is the > >> underlying reason for the calculation being so off from the mark? > > > > Two words: grid artefact! Basically, the deviation is not due to > > numerical problems, rather the high energy values you observe are the > > result of the selfinteraction of the grid points. That's why the > > deviation is increasing with higher resolution as the grid points are > > getting closer. What you have to do to get the total electrostatic > > energy of your system without selfenergies: > > > > 1) Compute the Coulomb energy in the homogeneous continuum. > > > > 2) Compute the solvation energy of the same charge distribution using > > APBS. The grid artefact cancels as you calculate an energy difference > > here. This, of course, requires that you use the same grid setup in both > > APBS runs. > > > > 3) Add the values together. > > > > Best regards, > > Gernot Kieseritzky > > > > >  > Return on Information: > Google Enterprise Search pays you back > Get the facts. > http://p.sf.net/sfu/googledev2dev > _______________________________________________ > apbsusers mailing list > apbsusers@... > https://lists.sourceforge.net/lists/listinfo/apbsusers >  Nathan Baker (http://bakergroup.wustl.edu) Associate Professor, Dept. of Biochemistry and Molecular Biophysics Director, Computational and Molecular Biophysics Graduate Program Center for Computational Biology, Washington University in St. Louis 
From: Gatti, Domenico <dgatti@me...>  20091211 15:54:10

Hi All, Could we post an EXAMPLE SCRIPT of how the corrections suggested by Gernot and Nathan for Dziedzic's simple calculation of two point charges should be implemented? Does this mean that the grid artifact subtraction should be carried out always in all the APBS calculations? Perhaps, this occurs already by default, and I did not realize it. Best, Domenico Domenico Gatti Biochemistry & Mol. Biology Wayne State University School of Medicine 540 E. Canfield Avenue Detroit, MI 48201 Tel: 3135770620 or 3139934238 Fax: 3135772765 dgatti@... Hi All  Gernot is absolutely correct. I would also add that, after correcting the issues Gernot raised below, you should also examine the sensitivity of your results on "chgm spl0" vs. "chgm spl2" since charge discretization can affect these types of calculations as well. Thanks, Nathan On Dec 10, 2009, at 9:31 AM, Gernot Kieseritzky wrote: > Hi! > > On Wed, 20091209 at 17:49 +0000, J.Dziedzic wrote: >> Hi! >> >> I am confused about the result of a very simple calculation when done >> with APBS. Consider a trivial system of two point charges with unit >> charges at a separation of 1 Bohr length, in vacuum. The Coulombic >> energy of this system is exactly 1 Hartree, that is 2625.5 kJ/mole. >> ... >> ... APBS yields 31740 kJ/mole, which off by a factor of 12. Making the >> grid finer only makes things worse, the results being: >> >> dime Energy (kJ/mole) >> 65 1.209E04 >> 129 2.222E04 >> 193 3.174E04 >> 257 4.058E04 >> 289 4.539E04 >> >> which are all way off, by a factor of 520, from the correct value of >> 2625.5 kJ/mole. >> >> I understand that normally one is interested in energy differences >> between a system in vacuo and a solvated system, and any discretization >> errors introduced are canceled if the grid is the same in both >> calculations. Yet, with a system so trivial, without any dielectric >> at all and, thus, without the arbitrariness of the cavity, what is the >> underlying reason for the calculation being so off from the mark? > > Two words: grid artefact! Basically, the deviation is not due to > numerical problems, rather the high energy values you observe are the > result of the selfinteraction of the grid points. That's why the > deviation is increasing with higher resolution as the grid points are > getting closer. What you have to do to get the total electrostatic > energy of your system without selfenergies: > > 1) Compute the Coulomb energy in the homogeneous continuum. > > 2) Compute the solvation energy of the same charge distribution using > APBS. The grid artefact cancels as you calculate an energy difference > here. This, of course, requires that you use the same grid setup in both > APBS runs. > > 3) Add the values together. > > Best regards, > Gernot Kieseritzky 
From: Nathan Baker <baker@bi...>  20091210 21:35:10

Hi All  Gernot is absolutely correct. I would also add that, after correcting the issues Gernot raised below, you should also examine the sensitivity of your results on "chgm spl0" vs. "chgm spl2" since charge discretization can affect these types of calculations as well. Thanks, Nathan On Dec 10, 2009, at 9:31 AM, Gernot Kieseritzky wrote: > Hi! > > On Wed, 20091209 at 17:49 +0000, J.Dziedzic wrote: >> Hi! >> >> I am confused about the result of a very simple calculation when done >> with APBS. Consider a trivial system of two point charges with unit >> charges at a separation of 1 Bohr length, in vacuum. The Coulombic >> energy of this system is exactly 1 Hartree, that is 2625.5 kJ/mole. >> ... >> ... APBS yields 31740 kJ/mole, which off by a factor of 12. Making the >> grid finer only makes things worse, the results being: >> >> dime Energy (kJ/mole) >> 65 1.209E04 >> 129 2.222E04 >> 193 3.174E04 >> 257 4.058E04 >> 289 4.539E04 >> >> which are all way off, by a factor of 520, from the correct value of >> 2625.5 kJ/mole. >> >> I understand that normally one is interested in energy differences >> between a system in vacuo and a solvated system, and any discretization >> errors introduced are canceled if the grid is the same in both >> calculations. Yet, with a system so trivial, without any dielectric >> at all and, thus, without the arbitrariness of the cavity, what is the >> underlying reason for the calculation being so off from the mark? > > Two words: grid artefact! Basically, the deviation is not due to > numerical problems, rather the high energy values you observe are the > result of the selfinteraction of the grid points. That's why the > deviation is increasing with higher resolution as the grid points are > getting closer. What you have to do to get the total electrostatic > energy of your system without selfenergies: > > 1) Compute the Coulomb energy in the homogeneous continuum. > > 2) Compute the solvation energy of the same charge distribution using > APBS. The grid artefact cancels as you calculate an energy difference > here. This, of course, requires that you use the same grid setup in both > APBS runs. > > 3) Add the values together. > > Best regards, > Gernot Kieseritzky > > >  > Return on Information: > Google Enterprise Search pays you back > Get the facts. > http://p.sf.net/sfu/googledev2dev > _______________________________________________ > apbsusers mailing list > apbsusers@... > https://lists.sourceforge.net/lists/listinfo/apbsusers  Nathan Baker (http://bakergroup.wustl.edu/) Associate Professor, Dept. of Biochemistry and Molecular Biophysics Director, Computational and Molecular Biophysics Graduate Program Center for Computational Biology, Washington University in St. Louis 
From: Gernot Kieseritzky <gernotf@ch...>  20091210 15:59:26

Hi! On Wed, 20091209 at 17:49 +0000, J.Dziedzic wrote: > Hi! > > I am confused about the result of a very simple calculation when done > with APBS. Consider a trivial system of two point charges with unit > charges at a separation of 1 Bohr length, in vacuum. The Coulombic > energy of this system is exactly 1 Hartree, that is 2625.5 kJ/mole. > ... > ... APBS yields 31740 kJ/mole, which off by a factor of 12. Making the > grid finer only makes things worse, the results being: > > dime Energy (kJ/mole) > 65 1.209E04 > 129 2.222E04 > 193 3.174E04 > 257 4.058E04 > 289 4.539E04 > > which are all way off, by a factor of 520, from the correct value of > 2625.5 kJ/mole. > > I understand that normally one is interested in energy differences > between a system in vacuo and a solvated system, and any discretization > errors introduced are canceled if the grid is the same in both > calculations. Yet, with a system so trivial, without any dielectric > at all and, thus, without the arbitrariness of the cavity, what is the > underlying reason for the calculation being so off from the mark? Two words: grid artefact! Basically, the deviation is not due to numerical problems, rather the high energy values you observe are the result of the selfinteraction of the grid points. That's why the deviation is increasing with higher resolution as the grid points are getting closer. What you have to do to get the total electrostatic energy of your system without selfenergies: 1) Compute the Coulomb energy in the homogeneous continuum. 2) Compute the solvation energy of the same charge distribution using APBS. The grid artefact cancels as you calculate an energy difference here. This, of course, requires that you use the same grid setup in both APBS runs. 3) Add the values together. Best regards, Gernot Kieseritzky 
From: J.Dziedzic <J.Dziedzic@so...>  20091209 17:49:48

Hi! I am confused about the result of a very simple calculation when done with APBS. Consider a trivial system of two point charges with unit charges at a separation of 1 Bohr length, in vacuum. The Coulombic energy of this system is exactly 1 Hartree, that is 2625.5 kJ/mole. When presented with the same system: REMARK 1 Two hydrogen atoms one Bohr apart ATOM 1 H XXX 1 0.000 0.000 0.000 1.00 1.00 ATOM 2 H XXX 1 0.529 0.000 0.000 1.00 1.00 and a simple input file: read mol pqr test.pqr end elec name test mgauto dime 193 193 193 nlev 4 cglen 50 50 50 fglen 12 12 12 cgcent mol 1 fgcent mol 1 mol 1 lpbe bcfl mdh pdie 1.0 sdie 1.0 chgm spl2 srfm smol swin 0.3 srad 1.4 sdens 10.0 temp 298.15 calcenergy total calcforce no end print elecEnergy test end quit ... APBS yields 31740 kJ/mole, which off by a factor of 12. Making the grid finer only makes things worse, the results being: dime Energy (kJ/mole) 65 1.209E04 129 2.222E04 193 3.174E04 257 4.058E04 289 4.539E04 which are all way off, by a factor of 520, from the correct value of 2625.5 kJ/mole. I understand that normally one is interested in energy differences between a system in vacuo and a solvated system, and any discretization errors introduced are canceled if the grid is the same in both calculations. Yet, with a system so trivial, without any dielectric at all and, thus, without the arbitrariness of the cavity, what is the underlying reason for the calculation being so off from the mark? Thank you in advance,  J. Dziedzic 
From: Nathan Baker <baker@bi...>  20091208 14:21:04

Hello  > I have several doubts regarding the details of an apbs calculations > that I hope you might answer. > > 1) Does inputting a charge density map invalidate the point charges > supplied in the pqr file or are _both_ the point charges and the > 'cloud' included in the calculation? I am assuming the latter. You are correct: only the cloud is included in the calculation. > 2) Does the electrostatic potential produced by 'write pot' include > the ions in any way or is it just due to the electron cloud (which is > what I need)? If it includes the ions, does it make sense to run > apbs with almostzero charges supplied in the pqr file to obtain the > potential due to the cloud only? It should be due to the cloud only. > 3) What is the rationale for having a set of three shifted diel maps? > If the grid spacing is very fine (0.075A), will supplying the same > map file for all three maps be a reasonable approximation or will > this make no sense whatsoever (I can imagine derivatives of diel > being wrong then). You are correct: the three shifted maps are needed for derivative calculations. > 4) Since the atomic cores are now embedded in a medium with a varying > dielectric constant, I can no longer use the Ewald technique in my > DFT code to compute the corecore interaction energy. Is there > any way I can obtain the corecore energy from apbs, i.e. the > energy of interaction of the point charges supplied in the pqr > file, embedded in a medium of varying diel, supplied as maps? > What about the corecloud energy, is there any way to obtain > this from apbs? Yes, you can set up a free energy cycle to transfer the core charges to a homogeneous dielectric medium (and calculate the transfer energy) and then calculate the Coulombic energy of the core point charges in the homogeneous dielectric. > 5) Is there any way to perform an apbs calculation with periodic > boundary conditions for the potential? Unfortunately, not yet. We're (still) working on this! Thanks, Nathan > Thank you in advance, >  J. Dziedzic > > > >  > Return on Information: > Google Enterprise Search pays you back > Get the facts. > http://p.sf.net/sfu/googledev2dev > _______________________________________________ > apbsusers mailing list > apbsusers@... > https://lists.sourceforge.net/lists/listinfo/apbsusers — Nathan Baker (http://bakergroup.wustl.edu/) Associate Professor, Dept. of Biochemistry and Molecular Biophysics Director, Computational and Molecular Biophysics Graduate Program Center for Computational Biology, Washington University in St. Louis 
From: J.Dziedzic <J.Dziedzic@so...>  20091207 11:06:55

Hello! I am trying to utilize the apbs code to include implicit solvent in a DFT calculation. My system is represented by an electronic charge density (the "cloud") sampled on a fine grid and a set of positive point charges representing the atomic cores. The positions of the cores are passed to apbs via the pqr file. The electronic charge density is passed to apbs via a charge map. A suitably defined map of smoothly varying dielectric constant is passed to apbs via a diel map (three maps, actually). Ionic concentrations are all zero. MDH boundary conditions are used. Manual focusing is utilized with progressively finer meshes, but the physical grid size is unchanged (since the potential in the whole box, not just in the vicinity of the molecule, is required as output  see below). Apbs is then used to calculate the resulting potential map, which is output to a .dx file. This map is then read by the DFT code, replacing the Hartree potential of the DFT calculation (the potential due to the electron cloud). A new charge distribution resulting from this potential is then computed, passed to apbs and the cycle continues until energy convergence is reached. I have several doubts regarding the details of an apbs calculations that I hope you might answer. 1) Does inputting a charge density map invalidate the point charges supplied in the pqr file or are _both_ the point charges and the 'cloud' included in the calculation? I am assuming the latter. 2) Does the electrostatic potential produced by 'write pot' include the ions in any way or is it just due to the electron cloud (which is what I need)? If it includes the ions, does it make sense to run apbs with almostzero charges supplied in the pqr file to obtain the potential due to the cloud only? 3) What is the rationale for having a set of three shifted diel maps? If the grid spacing is very fine (0.075A), will supplying the same map file for all three maps be a reasonable approximation or will this make no sense whatsoever (I can imagine derivatives of diel being wrong then). 4) Since the atomic cores are now embedded in a medium with a varying dielectric constant, I can no longer use the Ewald technique in my DFT code to compute the corecore interaction energy. Is there any way I can obtain the corecore energy from apbs, i.e. the energy of interaction of the point charges supplied in the pqr file, embedded in a medium of varying diel, supplied as maps? What about the corecloud energy, is there any way to obtain this from apbs? 5) Is there any way to perform an apbs calculation with periodic boundary conditions for the potential? Thank you in advance,  J. Dziedzic 
From: Nathan Baker <baker@bi...>  20091204 18:31:49

Hello  I would guess that your analysis and comparison would benefit from statistical tests to determine the significance of the difference in the presence of the errors you observe. This will allow you to answer your question about the reliability of your results. Good luck, Nathan On Dec 4, 2009, at 8:39 AM, Dimitrios Spiliotopoulos wrote: > > Dear prof Baker, > > the standard deviations for my calculations are now lower than before (around 30% of the mean value, very similar in absolute terms between the wildtype and the mutant complexes). > I calculated the DeltaDeltaG values (by that I define the difference of the mutant DeltaG and the wildtype DeltaG) of a number of mutants, obtaining a fairly good correlation. These standard deviations are nonetheless higher than the DeltaDeltaG values, thus possibly invaliding any correlation. > Does this nullify my results or are they still reliable? > > Thank you very much!!! > > Dimitrios Spiliotopoulos > > _________________________________________________________________________________________________ > Dulbecco Telethon Institute c/o DIBIT Scientific Institute > Biomolecular NMR Laboratory, 1B4 > Via Olgettina 58, 20132 Milano (Italy) > Tel : 00390226434348/5622/3497/4922 > Fax : 00390226434153 > Email : spiliotopoulos.dimitrios@...; dimitris3.16@... > Skype: dimitris3.16 > > > > > > > > > > 2009/7/30 Nathan Baker <baker@...> > Hi Dimitrios  > > Such variance is fairly standard in polar solvation energies. Part of the "problem" is related to sensitivity in the sharp dielectric boundary while part of the issue is related to the intrinsic conformational sensitivity of polar solvation. I'm not sure that such variance really is all that surprising. You could try a smoother dielectric boundary (e.g., splines) but please be aware that these require specialized parameters. > > Thanks, > > Nathan > > On Jul 29, 2009, at 10:12 AM, Dimitrios Spiliotopoulos wrote: > >> >> Hello APBS users! >> >> I have a question about the standard deviations values of my simulationderived structure files of a protein interacting with a peptide. >> >> I performed five simulations (each composed of 1 ns NVT equilibration and a 3 ns NPT production run) and pasted them together in order to get a single "overall" simulation. On this latter simulation, I performed the MM/PBSA calculations. >> I noticed that the polar solvation contribution had standard deviations that are 1 to 3fold the mean value (e.g., 36.8534 +/ 78.0839 kJ/mol or 88.9041 +/ 88.6444 kJ/mol), whereas my apolar solvation contribution has more reasonable results (the standard deviations are 35% of the mean values). I thought may be the conditions were too permissive and my simulation was oversampling, so I changed them into 100 ps NVT and 100 ps NPT. Thus, I calculated the values for all the frames (1000) or every 10 frames (100), but I had very similar results. >> >> I cannot figure out why the standard deviations of the polar solvation term are so high: is there any parameter in the polar solvation calculation input file that can affect this? >> >> Thank you very much in advance! >> >> Dimitrios Spiliotopoulos >> >> _________________________________________________________________________________________________ >> Dulbecco Telethon Institute c/o DIBIT Scientific Institute >> Biomolecular NMR Laboratory, 1B4 >> Via Olgettina 58, 20132 Milano (Italy) >> Tel : 00390226434348/5622/3497/4922 >> Fax : 00390226434153 >> Email : spiliotopoulos.dimitrios@...; dimitris3.16@... >> Skype: dimitris3.16 >> >>  >> Let Crystal Reports handle the reporting  Free Crystal Reports 2008 30Day >> trial. Simplify your report design, integration and deployment  and focus on >> what you do best, core application coding. Discover what's new with >> Crystal Reports now. http://p.sf.net/sfu/bobjjuly_______________________________________________ >> apbsusers mailing list >> apbsusers@... >> https://lists.sourceforge.net/lists/listinfo/apbsusers > > — > Associate Professor, Dept. of Biochemistry and Molecular Biophysics > Director, Computational and Molecular Biophysics Graduate Program > Center for Computational Biology, Washington University in St. Louis > Web: http://bakergroup.wustl.edu/ > > > > > > > > > > > > > > — Nathan Baker (http://bakergroup.wustl.edu/) Associate Professor, Dept. of Biochemistry and Molecular Biophysics Director, Computational and Molecular Biophysics Graduate Program Center for Computational Biology, Washington University in St. Louis 
From: Dimitrios Spiliotopoulos <dimitris3.16@gm...>  20091204 14:39:22

Dear prof Baker, the standard deviations for my calculations are now lower than before (around 30% of the mean value, very similar in absolute terms between the wildtype and the mutant complexes). I calculated the DeltaDeltaG values (by that I define the difference of the mutant DeltaG and the wildtype DeltaG) of a number of mutants, obtaining a fairly good correlation. These standard deviations are nonetheless higher than the DeltaDeltaG values, thus possibly invaliding any correlation. Does this nullify my results or are they still reliable? Thank you very much!!! Dimitrios Spiliotopoulos _________________________________________________________________________________________________ Dulbecco Telethon Institute c/o DIBIT Scientific Institute Biomolecular NMR Laboratory, 1B4 Via Olgettina 58, 20132 Milano (Italy) Tel : 00390226434348/5622/3497/4922 Fax : 00390226434153 Email : spiliotopoulos.dimitrios@...; dimitris3.16@... Skype: dimitris3.16 2009/7/30 Nathan Baker <baker@...> > Hi Dimitrios  > > Such variance is fairly standard in polar solvation energies. Part of the > "problem" is related to sensitivity in the sharp dielectric boundary while > part of the issue is related to the intrinsic conformational sensitivity of > polar solvation. I'm not sure that such variance really is all that > surprising. You could try a smoother dielectric boundary (e.g., splines) > but please be aware that these require specialized parameters. > > Thanks, > > Nathan > > On Jul 29, 2009, at 10:12 AM, Dimitrios Spiliotopoulos wrote: > > > Hello APBS users! > > I have a question about the standard deviations values of my > simulationderived structure files of a protein interacting with a peptide. > > I performed five simulations (each composed of 1 ns NVT equilibration and a > 3 ns NPT production run) and pasted them together in order to get a single > "overall" simulation. On this latter simulation, I performed the MM/PBSA > calculations. > I noticed that the polar solvation contribution had standard deviations > that are 1 to 3fold the mean value (e.g., 36.8534 +/ 78.0839 kJ/mol or > 88.9041 +/ 88.6444 kJ/mol), whereas my apolar solvation contribution has > more reasonable results (the standard deviations are 35% of the mean > values). I thought may be the conditions were too permissive and my > simulation was oversampling, so I changed them into 100 ps NVT and 100 ps > NPT. Thus, I calculated the values for all the frames (1000) or every 10 > frames (100), but I had very similar results. > > I cannot figure out why the standard deviations of the polar solvation term > are so high: is there any parameter in the polar solvation calculation input > file that can affect this? > > Thank you very much in advance! > > Dimitrios Spiliotopoulos > > > _________________________________________________________________________________________________ > Dulbecco Telethon Institute c/o DIBIT Scientific Institute > Biomolecular NMR Laboratory, 1B4 > Via Olgettina 58, 20132 Milano (Italy) > Tel : 00390226434348/5622/3497/4922 > Fax : 00390226434153 > Email : spiliotopoulos.dimitrios@...; dimitris3.16@... > Skype: dimitris3.16 > > >  > Let Crystal Reports handle the reporting  Free Crystal Reports 2008 30Day > > trial. Simplify your report design, integration and deployment  and focus > on > what you do best, core application coding. Discover what's new with > Crystal Reports now. > http://p.sf.net/sfu/bobjjuly_______________________________________________ > apbsusers mailing list > apbsusers@... > https://lists.sourceforge.net/lists/listinfo/apbsusers > > > — > Associate Professor, Dept. of Biochemistry and Molecular Biophysics > Director, Computational and Molecular Biophysics Graduate Program > Center for Computational Biology, Washington University in St. Louis > Web: http://bakergroup.wustl.edu/ > > > > > > > > > > > > > > 
From: Dimitrios Spiliotopoulos <dimitris3.16@gm...>  20091204 14:28:25

Dear Jason, dear prof Baker, thank you very much for your kind answers!!! d. _________________________________________________________________________________________________ Dulbecco Telethon Institute c/o DIBIT Scientific Institute Biomolecular NMR Laboratory, 1B4 Via Olgettina 58, 20132 Milano (Italy) Tel : 00390226434348/5622/3497/4922 Fax : 00390226434153 Email : spiliotopoulos.dimitrios@...; dimitris3.16@... Skype: dimitris3.16 
From: Nathan Baker <baker@bi...>  20091203 02:26:40

Dear APBS Users  APBS 1.2.1 has been released. This is a bugfix release specifically aimed at addressing a problem with certain nonlinear PoissonBoltzmann calculations. In particular, several users noticed that PoissonBoltzmann calculations which focused into a low dielectric region would occasionally generate very large (~1E25 kJ/mol) energies. This was due to a bug that caused instability in the nonlinear PoissonBoltzmann solver. This bug has been fixed in this release. The full list of changes is: • Added in warning into focusFillBound if there is a large value detected in setting the boundary conditions during a focusing calculation • Added in a check and abort in Vpmg_qmEnergy if chopped values are detected. This occurs under certain conditions for NPBE calculations where focusing cuts into a lowdielectric regions. • Fixed a bug in Vpmg_MolIon that causes npbe based calculations to return very large energies. This occurs under certain conditions for NPBE calculations where focusing cuts into a lowdielectric regions. More information about APBS as well as download links can be obtained from http://www.poissonboltzmann.org/. We apologize for this bug and thank you for your continued support of APBS. Sincerely, The APBS Development Team — Nathan Baker (http://bakergroup.wustl.edu/) Associate Professor, Dept. of Biochemistry and Molecular Biophysics Director, Computational and Molecular Biophysics Graduate Program Center for Computational Biology, Washington University in St. Louis 