## Re: [Apbs-users] Calculating membrane-protein binding energy

 Re: [Apbs-users] Calculating membrane-protein binding energy From: Cesar Luis Avila - 2007-02-26 23:42:04 ```I think it would be better to compare Delphi with APBS using the same system. I will try to run penta-Lys binding to 2:1 pcps membrane, using the same pdbs and parameters set. The dielectric map should be calculated nearly the same way by both programs. As soon as I have the results, I will post them. Regards Cesar Nathan Baker escribió: > Hi Cesar -- > > This is certainly a strange result. How did your setup with APBS > compare with the CHARMM calculation? How did you generate the > dielectric profile for APBS? > > Thanks, > > Nathan > > On Feb 25, 2007, at 4:23 PM, Cesar Luis Avila wrote: > > >> Well, I have tried to perform some of this calculations with APBS, >> using the some parameter set as those described in Ben Tal et al >> (1996), Murray et al (2002) (just to mention some examples). This >> parameter set where thoroughly tested with Delphi. >> I used two diferent approaches: >> ---------------------------------------------------------------------- >> ---------------------------------------- >> 1- Total Electrostatic energy of Complex - (protein + membrane). >> >> print energy 1 (complex) - 2 (protein) - 3 (membrane) end >> apbs0.log: Local net energy (PE 0) = -2.119044325924E+00 kJ/mol >> apbs1.log: Local net energy (PE 0) = -5.275141016003E+00 kJ/mol >> apbs2.log: Local net energy (PE 0) = -5.636772223256E+00 kJ/mol >> apbs3.log: Local net energy (PE 0) = 4.212956888514E+00 kJ/mol >> apbs4.log: Local net energy (PE 0) = -5.562750415241E+01 kJ/mol >> apbs5.log: Local net energy (PE 0) = -5.082387239822E+01 kJ/mol >> apbs6.log: Local net energy (PE 0) = -5.706839225518E+01 kJ/mol >> apbs7.log: Local net energy (PE 0) = -6.377577736030E+01 kJ/mol >> >> Global net energy = -2.36 E+02 kJ/mol (sum of all local net energy >> depicted above). >> Note: because the size of the system, I used parallel async >> calculation ( 2 x 2 x 2 processor array). >> The same grid was set for all calculations >> ---------------------------------------------------------------------- >> ---------------------------------------------------------------------- >> ----------- >> >> 2- Binding Free energy (as shown on APBS's tutorial) >> DGsolv (polar) complex = -2.14E+05 kJ/mol >> DGsolv (polar) protein = -4.61E+04 kJ/mol >> DGsolv (polar) membrane = -1.46E+05 kJ/mol >> >> DDGsolv (polar) = -2,201E+04 kJ/mol >> >> DGcoul complex = -3.62E+06 kJ/mol >> DGcoul protein = -1.34E+06 kJ/mol >> DGcoul membrane = -2.26E+06 kJ/mol >> >> DDGcoul = -2.34E+04 kJ/mol >> >> DDGbinding (polar + coul) = -4.54E+04 kJ/mol >> (Note that the apolar contribution to the free energy of binding is >> still missing). >> Different grids were set for complex, protein and membrane. As far >> as I understood the energy partitioning methods should be grid >> independent. >> ---------------------------------------------------------------------- >> ------------------------------------------------------------- >> I find that the energy calculated in either of both methods ( -2.36 >> E+02 kJ/mol and -4.54E+04 kJ/mol) is extremely high (the free >> energy of binding for model peptides is around -5kcal/mol, ie, -21 >> kJ/mol, as well as for several other periferic proteins). Using >> Charmm with an implicit membrane model, I have calculated a free >> energy of binding of about -2.8 kcal/mol. I was not able to compare >> with Delphi because of a segmentation fault when calculating this >> system. >> Best Regards >> Cesar Avila >> >> >> >> >> >> Nathan Baker escribió: >> >>> Hi Cesar -- >>> >>> We're working on adding some functionality to this end to APBS -- >>> however, it's still in the research stage. If the membrane is >>> explicit, then the binding calculations are straightforward. For >>> implicit membranes, you'll need to setup your own dielectric >>> maps. There's been some discussion on how to do this in the past >>> on this mailing list. >>> >>> Good luck! >>> >>> -- Nathan >>> >>> On Feb 21, 2007, at 8:20 AM, Cesar Luis Avila wrote: >>> >>> >>> >>>> Dear all, >>>> I would like to calculate the free energy of binding of protein >>>> membrane >>>> systems in a similar fashion to that described by Ben Tal et al >>>> (1996), >>>> Murray and Honig (2002). It is supposed that the protein is >>>> adsorbed to >>>> the membrane through electrostatic interactions, but do not >>>> significantly penetrate in it. Has anyone tried to perform this >>>> kind of >>>> calculus using APBS? Which would be the best way to calculate >>>> different >>>> contributions to the total binding energy (ie electrostatic, non- >>>> polar, >>>> etc) in APBS? >>>> Best Regards >>>> Cesar Avila >>>> >>>> -------------------------------------------------------------------- >>>> -- --- >>>> Take Surveys. Earn Cash. Influence the Future of IT >>>> Join SourceForge.net's Techsay panel and you'll get the chance >>>> to share your >>>> opinions on IT & business topics through brief surveys-and earn cash >>>> http://www.techsay.com/default.php? >>>> page=join.php&p=sourceforge&CID=DEVDEV >>>> _______________________________________________ >>>> apbs-users mailing list >>>> apbs-users@... >>>> https://lists.sourceforge.net/lists/listinfo/apbs-users >>>> >>>> >>> -- >>> Associate Professor, Dept. of Biochemistry and Molecular Biophysics >>> Center for Computational Biology, Washington University in St. Louis >>> Web: http://cholla.wustl.edu/ >>> >>> >>> >>> --------------------------------------------------------------------- >>> ---- >>> Take Surveys. Earn Cash. Influence the Future of IT >>> Join SourceForge.net's Techsay panel and you'll get the chance to >>> share your >>> opinions on IT & business topics through brief surveys-and earn cash >>> http://www.techsay.com/default.php? >>> page=join.php&p=sourceforge&CID=DEVDEV >>> _______________________________________________ >>> apbs-users mailing list >>> apbs-users@... >>> https://lists.sourceforge.net/lists/listinfo/apbs-users >>> >>> >>> > > -- > Associate Professor, Dept. of Biochemistry and Molecular Biophysics > Center for Computational Biology, Washington University in St. Louis > Web: http://cholla.wustl.edu/ > > > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share your > opinions on IT & business topics through brief surveys-and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > _______________________________________________ > apbs-users mailing list > apbs-users@... > https://lists.sourceforge.net/lists/listinfo/apbs-users > > ```

 [Apbs-users] Calculating membrane-protein binding energy From: Cesar Luis Avila - 2007-02-21 14:19:36 ```Dear all, I would like to calculate the free energy of binding of protein membrane systems in a similar fashion to that described by Ben Tal et al (1996), Murray and Honig (2002). It is supposed that the protein is adsorbed to the membrane through electrostatic interactions, but do not significantly penetrate in it. Has anyone tried to perform this kind of calculus using APBS? Which would be the best way to calculate different contributions to the total binding energy (ie electrostatic, non-polar, etc) in APBS? Best Regards Cesar Avila ```
 Re: [Apbs-users] Calculating membrane-protein binding energy From: Nathan Baker - 2007-02-24 13:41:14 ```Hi Cesar -- We're working on adding some functionality to this end to APBS -- however, it's still in the research stage. If the membrane is explicit, then the binding calculations are straightforward. For implicit membranes, you'll need to setup your own dielectric maps. There's been some discussion on how to do this in the past on this mailing list. Good luck! -- Nathan On Feb 21, 2007, at 8:20 AM, Cesar Luis Avila wrote: > Dear all, > I would like to calculate the free energy of binding of protein > membrane > systems in a similar fashion to that described by Ben Tal et al > (1996), > Murray and Honig (2002). It is supposed that the protein is > adsorbed to > the membrane through electrostatic interactions, but do not > significantly penetrate in it. Has anyone tried to perform this > kind of > calculus using APBS? Which would be the best way to calculate > different > contributions to the total binding energy (ie electrostatic, non- > polar, > etc) in APBS? > Best Regards > Cesar Avila > > ---------------------------------------------------------------------- > --- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to > share your > opinions on IT & business topics through brief surveys-and earn cash > http://www.techsay.com/default.php? > page=join.php&p=sourceforge&CID=DEVDEV > _______________________________________________ > apbs-users mailing list > apbs-users@... > https://lists.sourceforge.net/lists/listinfo/apbs-users -- Associate Professor, Dept. of Biochemistry and Molecular Biophysics Center for Computational Biology, Washington University in St. Louis Web: http://cholla.wustl.edu/ ```
 Re: [Apbs-users] Calculating membrane-protein binding energy From: Cesar Luis Avila - 2007-02-25 22:23:09 ```Well, I have tried to perform some of this calculations with APBS, using the some parameter set as those described in Ben Tal et al (1996), Murray et al (2002) (just to mention some examples). This parameter set where thoroughly tested with Delphi. I used two diferent approaches: -------------------------------------------------------------------------------------------------------------- 1- Total Electrostatic energy of Complex - (protein + membrane). print energy 1 (complex) - 2 (protein) - 3 (membrane) end apbs0.log: Local net energy (PE 0) = -2.119044325924E+00 kJ/mol apbs1.log: Local net energy (PE 0) = -5.275141016003E+00 kJ/mol apbs2.log: Local net energy (PE 0) = -5.636772223256E+00 kJ/mol apbs3.log: Local net energy (PE 0) = 4.212956888514E+00 kJ/mol apbs4.log: Local net energy (PE 0) = -5.562750415241E+01 kJ/mol apbs5.log: Local net energy (PE 0) = -5.082387239822E+01 kJ/mol apbs6.log: Local net energy (PE 0) = -5.706839225518E+01 kJ/mol apbs7.log: Local net energy (PE 0) = -6.377577736030E+01 kJ/mol Global net energy = -2.36 E+02 kJ/mol (sum of all local net energy depicted above). Note: because the size of the system, I used parallel async calculation ( 2 x 2 x 2 processor array). The same grid was set for all calculations ------------------------------------------------------------------------------------------------------------------------------------------------------- 2- Binding Free energy (as shown on APBS's tutorial) DGsolv (polar) complex = -2.14E+05 kJ/mol DGsolv (polar) protein = -4.61E+04 kJ/mol DGsolv (polar) membrane = -1.46E+05 kJ/mol DDGsolv (polar) = -2,201E+04 kJ/mol DGcoul complex = -3.62E+06 kJ/mol DGcoul protein = -1.34E+06 kJ/mol DGcoul membrane = -2.26E+06 kJ/mol DDGcoul = -2.34E+04 kJ/mol DDGbinding (polar + coul) = -4.54E+04 kJ/mol (Note that the apolar contribution to the free energy of binding is still missing). Different grids were set for complex, protein and membrane. As far as I understood the energy partitioning methods should be grid independent. ----------------------------------------------------------------------------------------------------------------------------------- I find that the energy calculated in either of both methods ( -2.36 E+02 kJ/mol and -4.54E+04 kJ/mol) is extremely high (the free energy of binding for model peptides is around -5kcal/mol, ie, -21 kJ/mol, as well as for several other periferic proteins). Using Charmm with an implicit membrane model, I have calculated a free energy of binding of about -2.8 kcal/mol. I was not able to compare with Delphi because of a segmentation fault when calculating this system. Best Regards Cesar Avila Nathan Baker escribió: > Hi Cesar -- > > We're working on adding some functionality to this end to APBS -- > however, it's still in the research stage. If the membrane is > explicit, then the binding calculations are straightforward. For > implicit membranes, you'll need to setup your own dielectric maps. > There's been some discussion on how to do this in the past on this > mailing list. > > Good luck! > > -- Nathan > > On Feb 21, 2007, at 8:20 AM, Cesar Luis Avila wrote: > > >> Dear all, >> I would like to calculate the free energy of binding of protein >> membrane >> systems in a similar fashion to that described by Ben Tal et al >> (1996), >> Murray and Honig (2002). It is supposed that the protein is >> adsorbed to >> the membrane through electrostatic interactions, but do not >> significantly penetrate in it. Has anyone tried to perform this >> kind of >> calculus using APBS? Which would be the best way to calculate >> different >> contributions to the total binding energy (ie electrostatic, non- >> polar, >> etc) in APBS? >> Best Regards >> Cesar Avila >> >> ---------------------------------------------------------------------- >> --- >> Take Surveys. Earn Cash. Influence the Future of IT >> Join SourceForge.net's Techsay panel and you'll get the chance to >> share your >> opinions on IT & business topics through brief surveys-and earn cash >> http://www.techsay.com/default.php? >> page=join.php&p=sourceforge&CID=DEVDEV >> _______________________________________________ >> apbs-users mailing list >> apbs-users@... >> https://lists.sourceforge.net/lists/listinfo/apbs-users >> > > -- > Associate Professor, Dept. of Biochemistry and Molecular Biophysics > Center for Computational Biology, Washington University in St. Louis > Web: http://cholla.wustl.edu/ > > > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share your > opinions on IT & business topics through brief surveys-and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > _______________________________________________ > apbs-users mailing list > apbs-users@... > https://lists.sourceforge.net/lists/listinfo/apbs-users > > ```
 Re: [Apbs-users] Calculating membrane-protein binding energy From: Nathan Baker - 2007-02-26 21:53:36 ```Hi Cesar -- This is certainly a strange result. How did your setup with APBS =20 compare with the CHARMM calculation? How did you generate the =20 dielectric profile for APBS? Thanks, Nathan On Feb 25, 2007, at 4:23 PM, Cesar Luis Avila wrote: > Well, I have tried to perform some of this calculations with APBS, =20 > using the some parameter set as those described in Ben Tal et al =20 > (1996), Murray et al (2002) (just to mention some examples). This =20 > parameter set where thoroughly tested with Delphi. > I used two diferent approaches: > ----------------------------------------------------------------------=20= > ---------------------------------------- > 1- Total Electrostatic energy of Complex - (protein + membrane). > > print energy 1 (complex) - 2 (protein) - 3 (membrane) end > apbs0.log: Local net energy (PE 0) =3D -2.119044325924E+00 kJ/mol > apbs1.log: Local net energy (PE 0) =3D -5.275141016003E+00 kJ/mol > apbs2.log: Local net energy (PE 0) =3D -5.636772223256E+00 kJ/mol > apbs3.log: Local net energy (PE 0) =3D 4.212956888514E+00 kJ/mol > apbs4.log: Local net energy (PE 0) =3D -5.562750415241E+01 kJ/mol > apbs5.log: Local net energy (PE 0) =3D -5.082387239822E+01 kJ/mol > apbs6.log: Local net energy (PE 0) =3D -5.706839225518E+01 kJ/mol > apbs7.log: Local net energy (PE 0) =3D -6.377577736030E+01 kJ/mol > > Global net energy =3D -2.36 E+02 kJ/mol (sum of all local net energy =20= > depicted above). > Note: because the size of the system, I used parallel async =20 > calculation ( 2 x 2 x 2 processor array). > The same grid was set for all calculations > ----------------------------------------------------------------------=20= > ----------------------------------------------------------------------=20= > ----------- > > 2- Binding Free energy (as shown on APBS's tutorial) > DGsolv (polar) complex =3D -2.14E+05 kJ/mol > DGsolv (polar) protein =3D -4.61E+04 kJ/mol > DGsolv (polar) membrane =3D -1.46E+05 kJ/mol > > DDGsolv (polar) =3D -2,201E+04 kJ/mol > > DGcoul complex =3D -3.62E+06 kJ/mol > DGcoul protein =3D -1.34E+06 kJ/mol > DGcoul membrane =3D -2.26E+06 kJ/mol > > DDGcoul =3D -2.34E+04 kJ/mol > > DDGbinding (polar + coul) =3D -4.54E+04 kJ/mol > (Note that the apolar contribution to the free energy of binding is =20= > still missing). > Different grids were set for complex, protein and membrane. As far =20 > as I understood the energy partitioning methods should be grid =20 > independent. > ----------------------------------------------------------------------=20= > ------------------------------------------------------------- > I find that the energy calculated in either of both methods ( -2.36 =20= > E+02 kJ/mol and -4.54E+04 kJ/mol) is extremely high (the free =20 > energy of binding for model peptides is around -5kcal/mol, ie, -21 =20 > kJ/mol, as well as for several other periferic proteins). Using =20 > Charmm with an implicit membrane model, I have calculated a free =20 > energy of binding of about -2.8 kcal/mol. I was not able to compare =20= > with Delphi because of a segmentation fault when calculating this =20 > system. > Best Regards > Cesar Avila > > > > > > Nathan Baker escribi=F3: >> Hi Cesar -- >> >> We're working on adding some functionality to this end to APBS -- =20= >> however, it's still in the research stage. If the membrane is =20 >> explicit, then the binding calculations are straightforward. For =20= >> implicit membranes, you'll need to setup your own dielectric =20 >> maps. There's been some discussion on how to do this in the past =20= >> on this mailing list. >> >> Good luck! >> >> -- Nathan >> >> On Feb 21, 2007, at 8:20 AM, Cesar Luis Avila wrote: >> >> >>> Dear all, >>> I would like to calculate the free energy of binding of protein =20 >>> membrane >>> systems in a similar fashion to that described by Ben Tal et al =20 >>> (1996), >>> Murray and Honig (2002). It is supposed that the protein is =20 >>> adsorbed to >>> the membrane through electrostatic interactions, but do not >>> significantly penetrate in it. Has anyone tried to perform this =20= >>> kind of >>> calculus using APBS? Which would be the best way to calculate =20 >>> different >>> contributions to the total binding energy (ie electrostatic, non- =20= >>> polar, >>> etc) in APBS? >>> Best Regards >>> Cesar Avila >>> >>> --------------------------------------------------------------------=20= >>> -- --- >>> Take Surveys. Earn Cash. Influence the Future of IT >>> Join SourceForge.net's Techsay panel and you'll get the chance =20 >>> to share your >>> opinions on IT & business topics through brief surveys-and earn cash >>> http://www.techsay.com/default.php? =20 >>> page=3Djoin.php&p=3Dsourceforge&CID=3DDEVDEV >>> _______________________________________________ >>> apbs-users mailing list >>> apbs-users@... >>> https://lists.sourceforge.net/lists/listinfo/apbs-users >>> >> >> -- >> Associate Professor, Dept. of Biochemistry and Molecular Biophysics >> Center for Computational Biology, Washington University in St. Louis >> Web: http://cholla.wustl.edu/ >> >> >> >> ---------------------------------------------------------------------=20= >> ---- >> Take Surveys. Earn Cash. Influence the Future of IT >> Join SourceForge.net's Techsay panel and you'll get the chance to =20 >> share your >> opinions on IT & business topics through brief surveys-and earn cash >> http://www.techsay.com/default.php?=20 >> page=3Djoin.php&p=3Dsourceforge&CID=3DDEVDEV >> _______________________________________________ >> apbs-users mailing list >> apbs-users@... >> https://lists.sourceforge.net/lists/listinfo/apbs-users >> >> > -- Associate Professor, Dept. of Biochemistry and Molecular Biophysics Center for Computational Biology, Washington University in St. Louis Web: http://cholla.wustl.edu/ ```
 Re: [Apbs-users] Calculating membrane-protein binding energy From: Cesar Luis Avila - 2007-02-26 23:42:04 ```I think it would be better to compare Delphi with APBS using the same system. I will try to run penta-Lys binding to 2:1 pcps membrane, using the same pdbs and parameters set. The dielectric map should be calculated nearly the same way by both programs. As soon as I have the results, I will post them. Regards Cesar Nathan Baker escribió: > Hi Cesar -- > > This is certainly a strange result. How did your setup with APBS > compare with the CHARMM calculation? How did you generate the > dielectric profile for APBS? > > Thanks, > > Nathan > > On Feb 25, 2007, at 4:23 PM, Cesar Luis Avila wrote: > > >> Well, I have tried to perform some of this calculations with APBS, >> using the some parameter set as those described in Ben Tal et al >> (1996), Murray et al (2002) (just to mention some examples). This >> parameter set where thoroughly tested with Delphi. >> I used two diferent approaches: >> ---------------------------------------------------------------------- >> ---------------------------------------- >> 1- Total Electrostatic energy of Complex - (protein + membrane). >> >> print energy 1 (complex) - 2 (protein) - 3 (membrane) end >> apbs0.log: Local net energy (PE 0) = -2.119044325924E+00 kJ/mol >> apbs1.log: Local net energy (PE 0) = -5.275141016003E+00 kJ/mol >> apbs2.log: Local net energy (PE 0) = -5.636772223256E+00 kJ/mol >> apbs3.log: Local net energy (PE 0) = 4.212956888514E+00 kJ/mol >> apbs4.log: Local net energy (PE 0) = -5.562750415241E+01 kJ/mol >> apbs5.log: Local net energy (PE 0) = -5.082387239822E+01 kJ/mol >> apbs6.log: Local net energy (PE 0) = -5.706839225518E+01 kJ/mol >> apbs7.log: Local net energy (PE 0) = -6.377577736030E+01 kJ/mol >> >> Global net energy = -2.36 E+02 kJ/mol (sum of all local net energy >> depicted above). >> Note: because the size of the system, I used parallel async >> calculation ( 2 x 2 x 2 processor array). >> The same grid was set for all calculations >> ---------------------------------------------------------------------- >> ---------------------------------------------------------------------- >> ----------- >> >> 2- Binding Free energy (as shown on APBS's tutorial) >> DGsolv (polar) complex = -2.14E+05 kJ/mol >> DGsolv (polar) protein = -4.61E+04 kJ/mol >> DGsolv (polar) membrane = -1.46E+05 kJ/mol >> >> DDGsolv (polar) = -2,201E+04 kJ/mol >> >> DGcoul complex = -3.62E+06 kJ/mol >> DGcoul protein = -1.34E+06 kJ/mol >> DGcoul membrane = -2.26E+06 kJ/mol >> >> DDGcoul = -2.34E+04 kJ/mol >> >> DDGbinding (polar + coul) = -4.54E+04 kJ/mol >> (Note that the apolar contribution to the free energy of binding is >> still missing). >> Different grids were set for complex, protein and membrane. As far >> as I understood the energy partitioning methods should be grid >> independent. >> ---------------------------------------------------------------------- >> ------------------------------------------------------------- >> I find that the energy calculated in either of both methods ( -2.36 >> E+02 kJ/mol and -4.54E+04 kJ/mol) is extremely high (the free >> energy of binding for model peptides is around -5kcal/mol, ie, -21 >> kJ/mol, as well as for several other periferic proteins). Using >> Charmm with an implicit membrane model, I have calculated a free >> energy of binding of about -2.8 kcal/mol. I was not able to compare >> with Delphi because of a segmentation fault when calculating this >> system. >> Best Regards >> Cesar Avila >> >> >> >> >> >> Nathan Baker escribió: >> >>> Hi Cesar -- >>> >>> We're working on adding some functionality to this end to APBS -- >>> however, it's still in the research stage. If the membrane is >>> explicit, then the binding calculations are straightforward. For >>> implicit membranes, you'll need to setup your own dielectric >>> maps. There's been some discussion on how to do this in the past >>> on this mailing list. >>> >>> Good luck! >>> >>> -- Nathan >>> >>> On Feb 21, 2007, at 8:20 AM, Cesar Luis Avila wrote: >>> >>> >>> >>>> Dear all, >>>> I would like to calculate the free energy of binding of protein >>>> membrane >>>> systems in a similar fashion to that described by Ben Tal et al >>>> (1996), >>>> Murray and Honig (2002). It is supposed that the protein is >>>> adsorbed to >>>> the membrane through electrostatic interactions, but do not >>>> significantly penetrate in it. Has anyone tried to perform this >>>> kind of >>>> calculus using APBS? Which would be the best way to calculate >>>> different >>>> contributions to the total binding energy (ie electrostatic, non- >>>> polar, >>>> etc) in APBS? >>>> Best Regards >>>> Cesar Avila >>>> >>>> -------------------------------------------------------------------- >>>> -- --- >>>> Take Surveys. Earn Cash. Influence the Future of IT >>>> Join SourceForge.net's Techsay panel and you'll get the chance >>>> to share your >>>> opinions on IT & business topics through brief surveys-and earn cash >>>> http://www.techsay.com/default.php? >>>> page=join.php&p=sourceforge&CID=DEVDEV >>>> _______________________________________________ >>>> apbs-users mailing list >>>> apbs-users@... >>>> https://lists.sourceforge.net/lists/listinfo/apbs-users >>>> >>>> >>> -- >>> Associate Professor, Dept. of Biochemistry and Molecular Biophysics >>> Center for Computational Biology, Washington University in St. Louis >>> Web: http://cholla.wustl.edu/ >>> >>> >>> >>> --------------------------------------------------------------------- >>> ---- >>> Take Surveys. Earn Cash. Influence the Future of IT >>> Join SourceForge.net's Techsay panel and you'll get the chance to >>> share your >>> opinions on IT & business topics through brief surveys-and earn cash >>> http://www.techsay.com/default.php? >>> page=join.php&p=sourceforge&CID=DEVDEV >>> _______________________________________________ >>> apbs-users mailing list >>> apbs-users@... >>> https://lists.sourceforge.net/lists/listinfo/apbs-users >>> >>> >>> > > -- > Associate Professor, Dept. of Biochemistry and Molecular Biophysics > Center for Computational Biology, Washington University in St. Louis > Web: http://cholla.wustl.edu/ > > > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share your > opinions on IT & business topics through brief surveys-and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > _______________________________________________ > apbs-users mailing list > apbs-users@... > https://lists.sourceforge.net/lists/listinfo/apbs-users > > ```
 Re: [Apbs-users] Calculating membrane-protein binding energy From: Nathan Baker - 2007-02-27 14:05:16 ```Hi Cesar -- APBS results have compared favorably to DelPhi, UHBD, and MEAD in the =20= past. If there are differences in the calculations you propose, =20 they'll likely be due to dielectric differences. It would be =20 particularly interesting to know the details of the dielectric =20 definitions and problem setup when you report the results of the APBS/=20= DelPhi comparison. Thanks, Nathan On Feb 26, 2007, at 5:42 PM, Cesar Luis Avila wrote: > I think it would be better to compare Delphi with APBS using the =20 > same system. I will try to run penta-Lys binding to 2:1 pcps =20 > membrane, using the same pdbs and parameters set. The dielectric =20 > map should be calculated nearly the same way by both programs. As =20 > soon as I have the results, I will post them. > > Regards > Cesar > > Nathan Baker escribi=F3: >> Hi Cesar -- >> >> This is certainly a strange result. How did your setup with APBS =20= >> compare with the CHARMM calculation? How did you generate the =20 >> dielectric profile for APBS? >> >> Thanks, >> >> Nathan >> >> On Feb 25, 2007, at 4:23 PM, Cesar Luis Avila wrote: >> >> >>> Well, I have tried to perform some of this calculations with =20 >>> APBS, using the some parameter set as those described in Ben Tal =20= >>> et al (1996), Murray et al (2002) (just to mention some =20 >>> examples). This parameter set where thoroughly tested with Delphi. >>> I used two diferent approaches: >>> --------------------------------------------------------------------=20= >>> -- ---------------------------------------- >>> 1- Total Electrostatic energy of Complex - (protein + membrane). >>> >>> print energy 1 (complex) - 2 (protein) - 3 (membrane) end >>> apbs0.log: Local net energy (PE 0) =3D -2.119044325924E+00 kJ/mol >>> apbs1.log: Local net energy (PE 0) =3D -5.275141016003E+00 kJ/mol >>> apbs2.log: Local net energy (PE 0) =3D -5.636772223256E+00 kJ/mol >>> apbs3.log: Local net energy (PE 0) =3D 4.212956888514E+00 kJ/mol >>> apbs4.log: Local net energy (PE 0) =3D -5.562750415241E+01 kJ/mol >>> apbs5.log: Local net energy (PE 0) =3D -5.082387239822E+01 kJ/mol >>> apbs6.log: Local net energy (PE 0) =3D -5.706839225518E+01 kJ/mol >>> apbs7.log: Local net energy (PE 0) =3D -6.377577736030E+01 kJ/mol >>> >>> Global net energy =3D -2.36 E+02 kJ/mol (sum of all local net =20 >>> energy depicted above). >>> Note: because the size of the system, I used parallel async =20 >>> calculation ( 2 x 2 x 2 processor array). >>> The same grid was set for all calculations >>> --------------------------------------------------------------------=20= >>> -- =20 >>> --------------------------------------------------------------------=20= >>> -- ----------- >>> >>> 2- Binding Free energy (as shown on APBS's tutorial) >>> DGsolv (polar) complex =3D -2.14E+05 kJ/mol >>> DGsolv (polar) protein =3D -4.61E+04 kJ/mol >>> DGsolv (polar) membrane =3D -1.46E+05 kJ/mol >>> >>> DDGsolv (polar) =3D -2,201E+04 kJ/mol >>> >>> DGcoul complex =3D -3.62E+06 kJ/mol >>> DGcoul protein =3D -1.34E+06 kJ/mol >>> DGcoul membrane =3D -2.26E+06 kJ/mol >>> >>> DDGcoul =3D -2.34E+04 kJ/mol >>> >>> DDGbinding (polar + coul) =3D -4.54E+04 kJ/mol >>> (Note that the apolar contribution to the free energy of binding =20 >>> is still missing). >>> Different grids were set for complex, protein and membrane. As =20 >>> far as I understood the energy partitioning methods should be =20 >>> grid independent. >>> --------------------------------------------------------------------=20= >>> -- ------------------------------------------------------------- >>> I find that the energy calculated in either of both methods =20 >>> ( -2.36 E+02 kJ/mol and -4.54E+04 kJ/mol) is extremely high (the =20= >>> free energy of binding for model peptides is around -5kcal/mol, =20 >>> ie, -21 kJ/mol, as well as for several other periferic =20 >>> proteins). Using Charmm with an implicit membrane model, I have =20 >>> calculated a free energy of binding of about -2.8 kcal/mol. I =20 >>> was not able to compare with Delphi because of a segmentation =20 >>> fault when calculating this system. >>> Best Regards >>> Cesar Avila >>> >>> >>> >>> >>> >>> Nathan Baker escribi=F3: >>> >>>> Hi Cesar -- >>>> >>>> We're working on adding some functionality to this end to APBS =20 >>>> -- however, it's still in the research stage. If the membrane =20= >>>> is explicit, then the binding calculations are =20 >>>> straightforward. For implicit membranes, you'll need to setup =20= >>>> your own dielectric maps. There's been some discussion on how =20= >>>> to do this in the past on this mailing list. >>>> >>>> Good luck! >>>> >>>> -- Nathan >>>> >>>> On Feb 21, 2007, at 8:20 AM, Cesar Luis Avila wrote: >>>> >>>> >>>> >>>>> Dear all, >>>>> I would like to calculate the free energy of binding of =20 >>>>> protein membrane >>>>> systems in a similar fashion to that described by Ben Tal et =20 >>>>> al (1996), >>>>> Murray and Honig (2002). It is supposed that the protein is =20 >>>>> adsorbed to >>>>> the membrane through electrostatic interactions, but do not >>>>> significantly penetrate in it. Has anyone tried to perform =20 >>>>> this kind of >>>>> calculus using APBS? Which would be the best way to calculate =20= >>>>> different >>>>> contributions to the total binding energy (ie electrostatic, =20 >>>>> non- polar, >>>>> etc) in APBS? >>>>> Best Regards >>>>> Cesar Avila >>>>> >>>>> ------------------------------------------------------------------=20= >>>>> -- -- --- >>>>> Take Surveys. Earn Cash. Influence the Future of IT >>>>> Join SourceForge.net's Techsay panel and you'll get the chance =20= >>>>> to share your >>>>> opinions on IT & business topics through brief surveys-and earn =20= >>>>> cash >>>>> http://www.techsay.com/default.php? =20 >>>>> page=3Djoin.php&p=3Dsourceforge&CID=3DDEVDEV >>>>> _______________________________________________ >>>>> apbs-users mailing list >>>>> apbs-users@... >>>>> https://lists.sourceforge.net/lists/listinfo/apbs-users >>>>> >>>>> >>>> -- >>>> Associate Professor, Dept. of Biochemistry and Molecular Biophysics >>>> Center for Computational Biology, Washington University in St. =20 >>>> Louis >>>> Web: http://cholla.wustl.edu/ >>>> >>>> >>>> >>>> -------------------------------------------------------------------=20= >>>> -- ---- >>>> Take Surveys. Earn Cash. Influence the Future of IT >>>> Join SourceForge.net's Techsay panel and you'll get the chance =20 >>>> to share your >>>> opinions on IT & business topics through brief surveys-and earn =20 >>>> cash >>>> http://www.techsay.com/default.php? =20 >>>> page=3Djoin.php&p=3Dsourceforge&CID=3DDEVDEV >>>> _______________________________________________ >>>> apbs-users mailing list >>>> apbs-users@... >>>> https://lists.sourceforge.net/lists/listinfo/apbs-users >>>> >>>> >>>> >> >> -- >> Associate Professor, Dept. of Biochemistry and Molecular Biophysics >> Center for Computational Biology, Washington University in St. Louis >> Web: http://cholla.wustl.edu/ >> >> >> >> ---------------------------------------------------------------------=20= >> ---- >> Take Surveys. Earn Cash. Influence the Future of IT >> Join SourceForge.net's Techsay panel and you'll get the chance to =20 >> share your >> opinions on IT & business topics through brief surveys-and earn cash >> http://www.techsay.com/default.php?=20 >> page=3Djoin.php&p=3Dsourceforge&CID=3DDEVDEV >> _______________________________________________ >> apbs-users mailing list >> apbs-users@... >> https://lists.sourceforge.net/lists/listinfo/apbs-users >> >> > -- Associate Professor, Dept. of Biochemistry and Molecular Biophysics Center for Computational Biology, Washington University in St. Louis Web: http://cholla.wustl.edu/ ```