Hello –

 

The total energies are often very susceptible to small changes in the charge distribution – these differences usually cancel out when calculating binding or solvation energies.

 

That being said, can you try running the calculations with “calcenergy comps” instead of “calcenergy total” so we can see which terms have the differences?

 

Thanks,

 

__________________________________________________
Nathan Baker
Pacific Northwest National Laboratory
Tel:  +1-509-375-3997

http://nabaker.me

 

From: Nuo Wang [mailto:wolich22@gmail.com]
Sent: Friday, August 09, 2013 10:06 AM
To: apbs-users@googlegroups.com
Cc: Nuo Wang; APBS-USERS mailing list (apbs-users@lists.sourceforge.net); Baker, Nathan
Subject: Re: [apbs-users] PQR charges vs. external charge map

 

Hi Dr. Baker,

I was testing the APBS's function by using a single lysine, PQR file as follows:

REMARK   1 PQR file generated by PDB2PQR (Version 1.8)
REMARK   1
REMARK   1 Forcefield Used: PARSE
REMARK   1
REMARK   5
REMARK   5 WARNING: PDB2PQR was unable to assign charges
REMARK   5          to the following atoms (omitted below):
REMARK   5              44 OXT in LYS 7
REMARK   5 This is usually due to the fact that this residue is not
REMARK   5 an amino acid or nucleic acid; or, there are no parameters
REMARK   5 available for the specific protonation state of this
REMARK   5 residue in the selected forcefield.
REMARK   5
REMARK   6 Total charge on this protein: 2.0000 e
REMARK   6
ATOM      1  N   LYS     7      86.648  21.188  64.073 -0.3200 2.0000
ATOM      2  CA  LYS     7      87.342  22.432  64.354  0.3300 2.0000
ATOM      3  C   LYS     7      86.395  23.417  65.020  0.5500 1.7000
ATOM      4  O   LYS     7      85.192  23.187  65.075 -0.5500 1.4000
ATOM      5  CB  LYS     7      87.922  23.006  63.060  0.0000 2.0000
ATOM      6  CG  LYS     7      88.938  22.060  62.413  0.0000 2.0000
ATOM      7  CD  LYS     7      89.643  22.651  61.187  0.0000 2.0000
ATOM      8  CE  LYS     7      88.699  22.853  59.995  0.3300 2.0000
ATOM      9  NZ  LYS     7      87.809  24.045  60.140 -0.3200 2.0000
ATOM     11  HA  LYS     7      88.107  22.248  64.979  0.0000 0.0000
ATOM     12  HE2 LYS     7      89.254  22.956  59.160  0.0000 0.0000
ATOM     13  HE3 LYS     7      88.130  22.027  59.893  0.0000 0.0000
ATOM     14  HG2 LYS     7      89.637  21.817  63.098  0.0000 0.0000
ATOM     15  HG3 LYS     7      88.464  21.216  62.134  0.0000 0.0000
ATOM     16  HZ1 LYS     7      87.829  24.575  59.294  0.3300 0.0000
ATOM     17  HZ3 LYS     7      86.877  23.738  60.323  0.3300 0.0000
ATOM     18  HZ2 LYS     7      88.134  24.607  60.898  0.3300 0.0000
ATOM     19  HD3 LYS     7      90.373  22.031  60.907  0.0000 0.0000
ATOM     20  HD2 LYS     7      90.028  23.537  61.436  0.0000 0.0000
ATOM     21  H2  LYS     7      86.860  20.510  64.782  0.3300 0.0000
ATOM     22  H3  LYS     7      85.656  21.345  64.057  0.3300 0.0000
ATOM     23  H   LYS     7      86.936  20.829  63.181  0.3300 0.0000
ATOM     24  HB3 LYS     7      88.382  23.856  63.272  0.0000 0.0000
ATOM     25  HB2 LYS     7      87.182  23.149  62.420  0.0000 0.0000
TER
END


First, I generate the charge map from this PQR file by write charge.
Second, I run APBS with the charge map generated above and the read charge and usemap commands (as shown in my input file).
In this second step, I tried to modify the PQR files slightly in different ways:
With just the PQR file shown above:

total electrostatic energy = 1.240938782423E+03 kJ/mol

With the first charge marked in red above changed to -9.9999:

total electrostatic energy = 1.217643377302E+03 kJ/mol

With the first charge marked in red above changed to 9.9999:

total electrostatic energy = 1.265773241262E+03 kJ/mol


We can see that as you add negative charge, the potential decreases slightly, as you add positive charge, the potential increases slightly. Also, to note that, if I change the first charge to -9.9999 and use the changed PQR file instead of the charge map (which contains the original LYS charges), the total electrostatic energy is 8.568383340586E+04 kJ/mol, which changed a fold!

So, I think, with the read charge and usemap, APBS does use the charge map provided, but somehow, the charges in the PQR file are still incorporated in some parameter in a minor way! I wonder why this is so?

Thanks a lot!
Nuo

On Thursday, August 8, 2013 8:27:01 PM UTC-7, Baker, Nathan wrote:

Hello –

 

Can you please provide more information about the specific comparison you are trying to perform? 

 

Thank you,

 

__________________________________________________
Nathan Baker
Pacific Northwest National Laboratory
Tel:  +1-509-375-3997
http://nabaker.me

 

From: Nuo Wang [mailto:woli...@gmail.com]
Sent: Thursday, August 08, 2013 7:08 PM
To: apbs-...@googlegroups.com
Subject: [apbs-users] PQR charges vs. external charge map

 

Hi APBS developers,

I'm trying to use an external charge map in APBS calculations, for example, my input file is:

read
    mol pqr protein.pqr
    charge dx external_charge_map.dx
end
elec
    mg-auto
    usemap charge 1
    dime 33 33 33
    cglen 13.3467 11.6569 14.8325
    fglen 13.3467 11.6569 14.8325
    cgcent mol 1
    fgcent mol 1
    mol 1
    lpbe
    bcfl sdh
    pdie 2.0000
    sdie 78.5400
    srfm smol
    chgm spl2
    sdens 10.00
    srad 1.40
    swin 0.30
    temp 298.15
    calcenergy total
    calcforce no
    write pot dx protein_pot
    write charge dx protein_charge
end
quit


As the documentation explained, with the usemap keyword, the fixed charges in the PQR file should be ignored. HOWEVER, I noticed that when I test to modify the charges in the PQR file (put all charges to 0, or add large charges), I end up getting slightly different total electrostatic energy displayed in the output. I wonder why this is so? If the PQR charges are not used as fixed charges, are they used somewhere else generating other parameters?

Thanks,
Nuo