Hydrogen bonding calculation issues with PDB2PQR/AMBER/CHIMERA
Hello- I’m trying to use CHIMERA to model the hexameric hydrogen bonding of isophthalic acid. I’m just an organic chemist, not a computational chemist. The PDB2PQR subroutine is returning the reply “Error encountered: No heavy atoms found!”. Here is sequentially what I have done: 1. A pdb file of isophthalic acid hexamer was imported. 2. The covalent bond lengths and angles were set to realistic ones. 3. The names of the concerned atoms were defined in the Build Structure/Modify Structure/Change Modified Residue’s name box, as described in the User Guide (i.e., Cac, O2, O3 and H) 4. The default limits of 3.4 Å and 30° were used 5. The AMBER force field was chosen. Thanks very much Fenton (PS – I’m a private person, not associated with either academia nor industry) -- Dr. Fenton Heirtzler Phone: +1-301-312-5145 On LinkedIn <http://www.linkedin.com/in/fentonhorgchem/> Skype: fentonheirtzler Supporter of the International Day of Climate Action <http://www.350.org/> *********
Hi Fenton, If you are just trying to detect the hydrogen bonds, I would definitely recommend using the FindHBond tool (in menu under Tools… Structure Analysis), or equivalently, the command “findhbond”. <http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/findhbond/findhbond.html> <http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/findhbond.html> If I understand correctly, PDB2PQR is not relevent to what you want to do. PDB2PQR uses a web service to assign partial charges to atoms by lookup table. Although its options include H-bonding stuff, that is only to allow it to add hydrogens if the structure doesn’t already have them. The different force field choices for PDB2PQR are different lookup tables, essentially, and the AMBER table only includes things like standard amino acids and nucleic acids, not your organic molecules. When the web service does not find the atoms in the lookup table, you get zero atoms in the results. <http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/apbs/pdb2pqr.html> Also, you do not need to change the names of the atoms. As long as the names imply the proper elements, the atom types will be detected from the bond lengths and angles. So theoretically, all you needed to do was read in your PDB file and use FindHBond. If the structure didn’t have hydrogens, FindHBond would still work (it would “imagine” where the hydrogens could be), or you could use AddH (in menu under Tools… Structure Editing) before running FindHBond. <http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/addh/addh.html> I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco On Aug 27, 2015, at 6:43 PM, Fenton Heirtzler <fentonh@gmail.com> wrote:
Hello- I’m trying to use CHIMERA to model the hexameric hydrogen bonding of isophthalic acid. I’m just an organic chemist, not a computational chemist. The PDB2PQR subroutine is returning the reply “Error encountered: No heavy atoms found!”. Here is sequentially what I have done:
1. A pdb file of isophthalic acid hexamer was imported.
2. The covalent bond lengths and angles were set to realistic ones.
3. The names of the concerned atoms were defined in the Build Structure/Modify Structure/Change Modified Residue’s name box, as described in the User Guide (i.e., Cac, O2, O3 and H)
4. The default limits of 3.4 Å and 30° were used
5. The AMBER force field was chosen.
Thanks very much Fenton
(PS – I’m a private person, not associated with either academia nor industry)
Hello Elaine, Thank you for your time. The isophthalic acid hexamer is, in fact already known. I want to reproduce the hydrogen bonds in Chimera and which were reported for its X-ray crystal structure, and not analyze a new X-ray structure. Hence "detecting" hydrogen bonds is not the goal. Afterwards, I want to use Chimera to predict the structure of a much more complicated compound which also contains the isophthalic acid hexamer motif. It was eventually possible to get Chimera to work for the simple isophthalic acid hexamer. But the more complicated structure is making problems, and that is why I am now writing you. Here is what ended up working for the "simple" isophthalic acid hexamer : a. Assigning all atoms (as you pointed out, this may or may not have helped) b. geometrically arrange the 6 molecules' O-H bonds so that they point at each other for H -bonding. c. Add H’s: yes; Add charges: yes (even though the structure already has all hydrogen atoms and has no charge) d. Use AM1-BCC , selecting the H-bonding option. However, I then tried the same procedure on the more complicated derivative. The following message appeared "The bond number exceeds the MAXBOND"....and Chimera then crashed. The more complicated molecule has 480 atoms & 522 bonds in total. 170 of the atoms are Hs attached to carbon : I tried "hiding" them, and it still crashed. The "MAXBOND" descriptor is confusing: the AMBER manual refers once to it as (maybe) a maximum bond length. But then the following link <http://svn.cgl.ucsf.edu/svn/chimera/trunk/foreign/antechamber/antechamber_1.27.patch> suggests that it is a number of bonds. And there is no mention of this parameter in the Chimera user manual. I just want to press a button somewhere and enter in a new value so that Chimera will not crash. Do you know if this is possible, or is my file just too large for the program? Thanks again On Fri, Aug 28, 2015 at 10:52 AM, Elaine Meng <meng@cgl.ucsf.edu> wrote:
Hi Fenton, If you are just trying to detect the hydrogen bonds, I would definitely
recommend using the FindHBond tool (in menu under Tools… Structure Analysis), or equivalently, the command “findhbond”.
< http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/findhbond/findhbon...
<http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/findhbond.html>
If I understand correctly, PDB2PQR is not relevent to what you want to do. PDB2PQR uses a web service to assign partial charges to atoms by lookup table. Although its options include H-bonding stuff, that is only to allow it to add hydrogens if the structure doesn’t already have them. The different force field choices for PDB2PQR are different lookup tables, essentially, and the AMBER table only includes things like standard amino acids and nucleic acids, not your organic molecules. When the web service does not find the atoms in the lookup table, you get zero atoms in the results. < http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/apbs/pdb2pqr.html>
Also, you do not need to change the names of the atoms. As long as the names imply the proper elements, the atom types will be detected from the bond lengths and angles.
So theoretically, all you needed to do was read in your PDB file and use FindHBond. If the structure didn’t have hydrogens, FindHBond would still work (it would “imagine” where the hydrogens could be), or you could use AddH (in menu under Tools… Structure Editing) before running FindHBond. <http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/addh/addh.html>
I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco
On Aug 27, 2015, at 6:43 PM, Fenton Heirtzler <fentonh@gmail.com> wrote:
Hello- I’m trying to use CHIMERA to model the hexameric hydrogen bonding of isophthalic acid. I’m just an organic chemist, not a computational chemist. The PDB2PQR subroutine is returning the reply “Error encountered: No heavy atoms found!”. Here is sequentially what I have done:
1. A pdb file of isophthalic acid hexamer was imported.
2. The covalent bond lengths and angles were set to realistic ones.
3. The names of the concerned atoms were defined in the Build Structure/Modify Structure/Change Modified Residue’s name box, as described in the User Guide (i.e., Cac, O2, O3 and H)
4. The default limits of 3.4 Å and 30° were used
5. The AMBER force field was chosen.
Thanks very much Fenton
(PS – I’m a private person, not associated with either academia nor industry)
-- Dr. Fenton Heirtzler Phone: +1-301-312-5145 On LinkedIn Skype: fentonheirtzler Supporter of the International Day of Climate Action *********
Hi Fenton, It sounds like you are trying to energy-minimize the structure. Note this is not “predicting” the structure of a molecule, because it will just go to the local minimum energy well most accessible from the starting structure, which usually isn’t the global minimum. Depending on the ruggedness of the energy surface, you could get a different result from each different starting conformation of the same molecule. However, putting that aside, let’s say you are trying to energy-minimize. You didn’t say that in your messages, but that’s my best guess since it will call other tools (AddH, AddCharge) to parametrize the structure beforehand. There is no need to say “yes” to adding hydrogens if your structure already has its hydrogens. It’s not going to do anything in that case. Conversely, it is necessary to add atomic charges before energy-minimization. Even if the molecule has no net charge, the individual atoms have partial charges. There is no H-bonding option to AM1-BCC charge calculation, it’s only an option for hydrogen addition (which was not done because the molecule already had hydrogens). MAXBOND is not something you can change in the user interface. I believe it is a limit on the molecule size that is set in the AMBER code, and the message means that your molecule has too many bonds for this routine to handle. You can choose the other charge calculation method (Gasteiger) instead of AM1-BCC. It is simpler and more approximate but it may be your only choice for this molecule. After the charges are added, the minimization can run. I hope this clarifies things, Elaine ---------- Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco
On Aug 28, 2015, at 1:13 PM, Fenton Heirtzler <fentonh@gmail.com> wrote:
Hello Elaine,
Thank you for your time. The isophthalic acid hexamer is, in fact already known. I want to reproduce the hydrogen bonds in Chimera and which were reported for its X-ray crystal structure, and not analyze a new X-ray structure. Hence "detecting" hydrogen bonds is not the goal. Afterwards, I want to use Chimera to predict the structure of a much more complicated compound which also contains the isophthalic acid hexamer motif.
It was eventually possible to get Chimera to work for the simple isophthalic acid hexamer. But the more complicated structure is making problems, and that is why I am now writing you.
Here is what ended up working for the "simple" isophthalic acid hexamer: a. Assigning all atoms (as you pointed out, this may or may not have helped) b. geometrically arrange the 6 molecules' O-H bonds so that they point at each other for H-bonding. c. Add H’s: yes; Add charges: yes (even though the structure already has all hydrogen atoms and has no charge) d. Use AM1-BCC, selecting the H-bonding option.
However, I then tried the same procedure on the more complicated derivative. The following message appeared "The bond number exceeds the MAXBOND"....and Chimera then crashed. The more complicated molecule has 480 atoms & 522 bonds in total. 170 of the atoms are Hs attached to carbon: I tried "hiding" them, and it still crashed.
The "MAXBOND" descriptor is confusing: the AMBER manual refers once to it as (maybe) a maximum bond length. But then the following link suggests that it is a number of bonds. And there is no mention of this parameter in the Chimera user manual. I just want to press a button somewhere and enter in a new value so that Chimera will not crash. Do you know if this is possible, or is my file just too large for the program?
Thanks again
On Fri, Aug 28, 2015 at 10:52 AM, Elaine Meng <meng@cgl.ucsf.edu> wrote:
Hi Fenton, If you are just trying to detect the hydrogen bonds, I would definitely recommend using the FindHBond tool (in menu under Tools… Structure Analysis), or equivalently, the command “findhbond”. <http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/findhbond/findhbond.html> <http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/findhbond.html>
If I understand correctly, PDB2PQR is not relevent to what you want to do. PDB2PQR uses a web service to assign partial charges to atoms by lookup table. Although its options include H-bonding stuff, that is only to allow it to add hydrogens if the structure doesn’t already have them. The different force field choices for PDB2PQR are different lookup tables, essentially, and the AMBER table only includes things like standard amino acids and nucleic acids, not your organic molecules. When the web service does not find the atoms in the lookup table, you get zero atoms in the results. <http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/apbs/pdb2pqr.html>
Also, you do not need to change the names of the atoms. As long as the names imply the proper elements, the atom types will be detected from the bond lengths and angles.
So theoretically, all you needed to do was read in your PDB file and use FindHBond. If the structure didn’t have hydrogens, FindHBond would still work (it would “imagine” where the hydrogens could be), or you could use AddH (in menu under Tools… Structure Editing) before running FindHBond. <http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/addh/addh.html>
I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco
On Aug 27, 2015, at 6:43 PM, Fenton Heirtzler <fentonh@gmail.com> wrote:
Hello- I’m trying to use CHIMERA to model the hexameric hydrogen bonding of isophthalic acid. I’m just an organic chemist, not a computational chemist. The PDB2PQR subroutine is returning the reply “Error encountered: No heavy atoms found!”. Here is sequentially what I have done:
1. A pdb file of isophthalic acid hexamer was imported.
2. The covalent bond lengths and angles were set to realistic ones.
3. The names of the concerned atoms were defined in the Build Structure/Modify Structure/Change Modified Residue’s name box, as described in the User Guide (i.e., Cac, O2, O3 and H)
4. The default limits of 3.4 Å and 30° were used
5. The AMBER force field was chosen.
Thanks very much Fenton
(PS – I’m a private person, not associated with either academia nor industry)
-- Dr. Fenton Heirtzler Phone: +1-301-312-5145 On LinkedIn Skype: fentonheirtzler Supporter of the International Day of Climate Action ********* _______________________________________________ Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
Hi Elaine, Thanks again for your time. Your explanation was very helpful. Best Fenton On Mon, Aug 31, 2015 at 3:28 PM, Elaine Meng <meng@cgl.ucsf.edu> wrote:
Hi Fenton, It sounds like you are trying to energy-minimize the structure. Note this is not “predicting” the structure of a molecule, because it will just go to the local minimum energy well most accessible from the starting structure, which usually isn’t the global minimum. Depending on the ruggedness of the energy surface, you could get a different result from each different starting conformation of the same molecule.
However, putting that aside, let’s say you are trying to energy-minimize. You didn’t say that in your messages, but that’s my best guess since it will call other tools (AddH, AddCharge) to parametrize the structure beforehand.
There is no need to say “yes” to adding hydrogens if your structure already has its hydrogens. It’s not going to do anything in that case.
Conversely, it is necessary to add atomic charges before energy-minimization. Even if the molecule has no net charge, the individual atoms have partial charges.
There is no H-bonding option to AM1-BCC charge calculation, it’s only an option for hydrogen addition (which was not done because the molecule already had hydrogens). MAXBOND is not something you can change in the user interface. I believe it is a limit on the molecule size that is set in the AMBER code, and the message means that your molecule has too many bonds for this routine to handle. You can choose the other charge calculation method (Gasteiger) instead of AM1-BCC. It is simpler and more approximate but it may be your only choice for this molecule.
After the charges are added, the minimization can run.
I hope this clarifies things, Elaine ---------- Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco
On Aug 28, 2015, at 1:13 PM, Fenton Heirtzler <fentonh@gmail.com> wrote:
Hello Elaine,
Thank you for your time. The isophthalic acid hexamer is, in fact already known. I want to reproduce the hydrogen bonds in Chimera and which were reported for its X-ray crystal structure, and not analyze a new X-ray structure. Hence "detecting" hydrogen bonds is not the goal. Afterwards, I want to use Chimera to predict the structure of a much more complicated compound which also contains the isophthalic acid hexamer motif.
It was eventually possible to get Chimera to work for the simple isophthalic acid hexamer. But the more complicated structure is making problems, and that is why I am now writing you.
Here is what ended up working for the "simple" isophthalic acid hexamer: a. Assigning all atoms (as you pointed out, this may or may not have helped) b. geometrically arrange the 6 molecules' O-H bonds so that they point at each other for H-bonding. c. Add H’s: yes; Add charges: yes (even though the structure already has all hydrogen atoms and has no charge) d. Use AM1-BCC, selecting the H-bonding option.
However, I then tried the same procedure on the more complicated derivative. The following message appeared "The bond number exceeds the MAXBOND"....and Chimera then crashed. The more complicated molecule has 480 atoms & 522 bonds in total. 170 of the atoms are Hs attached to carbon: I tried "hiding" them, and it still crashed.
The "MAXBOND" descriptor is confusing: the AMBER manual refers once to it as (maybe) a maximum bond length. But then the following link suggests that it is a number of bonds. And there is no mention of this parameter in the Chimera user manual. I just want to press a button somewhere and enter in a new value so that Chimera will not crash. Do you know if this is possible, or is my file just too large for the program?
Thanks again
On Fri, Aug 28, 2015 at 10:52 AM, Elaine Meng <meng@cgl.ucsf.edu> wrote:
Hi Fenton, If you are just trying to detect the hydrogen bonds, I would
definitely recommend using the FindHBond tool (in menu under Tools… Structure Analysis), or equivalently, the command “findhbond”.
< http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/findhbond/findhbon...
<http://www.rbvi.ucsf.edu/chimera/docs/UsersGuide/midas/findhbond.html
If I understand correctly, PDB2PQR is not relevent to what you want to
do. PDB2PQR uses a web service to assign partial charges to atoms by lookup table. Although its options include H-bonding stuff, that is only to allow it to add hydrogens if the structure doesn’t already have them. The different force field choices for PDB2PQR are different lookup tables, essentially, and the AMBER table only includes things like standard amino acids and nucleic acids, not your organic molecules. When the web service does not find the atoms in the lookup table, you get zero atoms in the results.
< http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/apbs/pdb2pqr.html
Also, you do not need to change the names of the atoms. As long as
the names imply the proper elements, the atom types will be detected from the bond lengths and angles.
So theoretically, all you needed to do was read in your PDB file and
use FindHBond. If the structure didn’t have hydrogens, FindHBond would still work (it would “imagine” where the hydrogens could be), or you could use AddH (in menu under Tools… Structure Editing) before running FindHBond.
< http://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/addh/addh.html>
I hope this helps, Elaine ----- Elaine C. Meng, Ph.D. UCSF Computer Graphics Lab (Chimera team) and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco
On Aug 27, 2015, at 6:43 PM, Fenton Heirtzler <fentonh@gmail.com> wrote:
Hello- I’m trying to use CHIMERA to model the hexameric hydrogen bonding of isophthalic acid. I’m just an organic chemist, not a computational chemist. The PDB2PQR subroutine is returning the reply “Error encountered: No heavy atoms found!”. Here is sequentially what I have done:
1. A pdb file of isophthalic acid hexamer was imported.
2. The covalent bond lengths and angles were set to realistic ones.
3. The names of the concerned atoms were defined in the Build Structure/Modify Structure/Change Modified Residue’s name box, as described in the User Guide (i.e., Cac, O2, O3 and H)
4. The default limits of 3.4 Å and 30° were used
5. The AMBER force field was chosen.
Thanks very much Fenton
(PS – I’m a private person, not associated with either academia nor industry)
-- Dr. Fenton Heirtzler Phone: +1-301-312-5145 On LinkedIn Skype: fentonheirtzler Supporter of the International Day of Climate Action ********* _______________________________________________ Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
-- Dr. Fenton Heirtzler Phone: +1-301-312-5145 On LinkedIn <http://www.linkedin.com/in/fentonhorgchem/> Skype: fentonheirtzler Supporter of the International Day of Climate Action <http://www.350.org/> *********
participants (2)
-
Elaine Meng -
Fenton Heirtzler