ffAMBER FAQ
AMBER force field ports for the GROMACS molecular dynamics suite
Maintained by Eric J. Sorin, Ph.D.
Department of Chemistry & Biochemistry
California State University, Long Beach
CONTENTS:
| General: | Can I use these
ports with any version of GROMACS? Do these ports support dummy atoms? Why do all ports include atom names like either amber94_XX or amber99_XX? How do I add new molecules to these ports? How do I convert AMBER topologies to GROMACS file formats? |
| pdb2gmx: | Can I import pdb
files with some or all hydrogens missing (-ignh
flag)? Why do LYS and CYS residues not follow standard residue nomenclature? Why does pdb2gmx keep trying to find "LYSH"? Why does pdb2gmx try to replace "LYS" with "LYN"? Why does pdb2gmx give me a .top file with more dihedrals than the one you have posted? |
| grompp: | Why do I get a
grompp warning about missing ' characters? Why do I get a grompp warning "missing white space after `#define proper_*'?" Why do I get warnings such as "No default [bond/angle/dihedral] type?" Why does my system have a non-integer total charge? |
| Other: | Why are there
non-zero energy differences between the GROMACS
ports and the AMBER 8.0 package? Why do I observe
slight energy differences between this GROMACS
port and the AMBER package for a system
containing TRP residues? |
General:
Can I use these ports with any version of GROMACS?
No! The ports are numbered to coincide with GROMACS releases, and minor differences between GROMACS versions render the ports non-interchangable.
Do these ports support dummy atoms?
No. Dummy hydrogen parameters for proteins are not yet complete and dummy database (.ddb) files are not provided. We hope to have this feature implemented in the next version of the AMBER ports.
Why do all ports include atom names like either amber94_XX or amber99_XX?
The other ports are all variants of AMBER-94 (96, GS, GS-S) or AMBER-99 (99phi), and they were created from the validated AMBER-94 and AMBER-99 files respectively. These variants only change peptide phi/psi torsions, and it is convenient to maintain the atom names.
How do I add new molecules to these ports?
While we used to provide a script to convert AMBER topology files into GROMACS format, that script has been replaced by ACPYPE - AnteChamber PYthon Parser interfacE, which users should find quite useful!
How do I convert AMBER topologies to GROMACS file formats?
While we used to provide a script to convert AMBER topology files into GROMACS format, that script has been replaced by ACPYPE - AnteChamber PYthon Parser interfacE, which users should find quite useful!
Can I use these ports with any version of GROMACS?
No! The ports are numbered to coincide with GROMACS releases, and minor differences between GROMACS versions render the ports non-interchangable.
Do these ports support dummy atoms?
No. Dummy hydrogen parameters for proteins are not yet complete and dummy database (.ddb) files are not provided. We hope to have this feature implemented in the next version of the AMBER ports.
Why do all ports include atom names like either amber94_XX or amber99_XX?
The other ports are all variants of AMBER-94 (96, GS, GS-S) or AMBER-99 (99phi), and they were created from the validated AMBER-94 and AMBER-99 files respectively. These variants only change peptide phi/psi torsions, and it is convenient to maintain the atom names.
How do I add new molecules to these ports?
While we used to provide a script to convert AMBER topology files into GROMACS format, that script has been replaced by ACPYPE - AnteChamber PYthon Parser interfacE, which users should find quite useful!
How do I convert AMBER topologies to GROMACS file formats?
While we used to provide a script to convert AMBER topology files into GROMACS format, that script has been replaced by ACPYPE - AnteChamber PYthon Parser interfacE, which users should find quite useful!
pdb2gmx:
Can I import pdb files with some or all hydrogens missing (-ignh flag)?
Yes and No. Protein ports have working hydrogen database (.hdb) files in all versions of our ports, so it is not necessary to have all-atom pdb models (i.e. including hydrogens). It is essential to copy aminoacids.dat to the top directory for this feature to work properly. Hydrogen databases are also present for nucleic acids in GROMACS versions 3.3 ports, but this feature was not plausible in previous versions of pdb2gmx, which makes 3.3 a nice improvement for those simulating nucleic acids. For nucleic acid simulations in versions 3.1.4 and 3.2.1 all-atom pdb models are still needed (i.e. united atom models will need to be protonated ahead of time using outside software).
Why do LYS and CYS residues not follow standard residue nomenclature?
Why does pdb2gmx keep trying to find "LYSH"?
Why does pdb2gmx try to replace "LYS" with "LYN"?
This is due to a substring comparison in pdb2gmx that automatically renames any residues with names containing "LYS" or "CYS" with the "LYSH" and "CYS" nomers. Because of this hard coding, we cannot offer a simple work around to maintain the AMBER residue nomenclaure. As discussed on the website, renaming LYS residues to LYP (Lysine positive) in your .pdb file will get rid of these warnings and keep pdb2gmx from trying to use LYN (unprotonated Lysine neutral).
Why does pdb2gmx give me a .top file with more dihedrals than the one you have posted?
Most likely, you are using an updated version of GROMACS: newer versions include all torsion terms in the topology even if the coefficients of these functions are all zero..
Can I import pdb files with some or all hydrogens missing (-ignh flag)?
Yes and No. Protein ports have working hydrogen database (.hdb) files in all versions of our ports, so it is not necessary to have all-atom pdb models (i.e. including hydrogens). It is essential to copy aminoacids.dat to the top directory for this feature to work properly. Hydrogen databases are also present for nucleic acids in GROMACS versions 3.3 ports, but this feature was not plausible in previous versions of pdb2gmx, which makes 3.3 a nice improvement for those simulating nucleic acids. For nucleic acid simulations in versions 3.1.4 and 3.2.1 all-atom pdb models are still needed (i.e. united atom models will need to be protonated ahead of time using outside software).
Why do LYS and CYS residues not follow standard residue nomenclature?
Why does pdb2gmx keep trying to find "LYSH"?
Why does pdb2gmx try to replace "LYS" with "LYN"?
This is due to a substring comparison in pdb2gmx that automatically renames any residues with names containing "LYS" or "CYS" with the "LYSH" and "CYS" nomers. Because of this hard coding, we cannot offer a simple work around to maintain the AMBER residue nomenclaure. As discussed on the website, renaming LYS residues to LYP (Lysine positive) in your .pdb file will get rid of these warnings and keep pdb2gmx from trying to use LYN (unprotonated Lysine neutral).
Why does pdb2gmx give me a .top file with more dihedrals than the one you have posted?
Most likely, you are using an updated version of GROMACS: newer versions include all torsion terms in the topology even if the coefficients of these functions are all zero..
grompp:
Why do I get a warning about missing ' characters?
You can set "cpp = /lib/cpp -traditional" in your .mdp file to get rid of warnings about missing ' characters, which is due to the use of proper nucleic acid nomenclature, such as to distinguish between C4' and C4.
Why do I get a grompp warning "missing white space after `#define proper_*'?"
Set "cpp = /lib/cpp -traditional" in your .mdp file to remove these warnings, which result from the use of the N* atom type of the AMBER family of force fields and can be ignored.
Why do I get warnings such as "No default [bond/angle/dihedral] type?"
These warnings may show up in some circumstances and are due to the many zero-valued parameters in AMBER force fields. Ignore any such warnings if you are certain that your force field files have not been modified since downloading from the ffAMBER website.
Why does my system have a non-integer total charge?
See the sections titled Important notes on using the AMBER ports.
Why do I get a warning about missing ' characters?
You can set "cpp = /lib/cpp -traditional" in your .mdp file to get rid of warnings about missing ' characters, which is due to the use of proper nucleic acid nomenclature, such as to distinguish between C4' and C4.
Why do I get a grompp warning "missing white space after `#define proper_*'?"
Set "cpp = /lib/cpp -traditional" in your .mdp file to remove these warnings, which result from the use of the N* atom type of the AMBER family of force fields and can be ignored.
Why do I get warnings such as "No default [bond/angle/dihedral] type?"
These warnings may show up in some circumstances and are due to the many zero-valued parameters in AMBER force fields. Ignore any such warnings if you are certain that your force field files have not been modified since downloading from the ffAMBER website.
Why does my system have a non-integer total charge?
See the sections titled Important notes on using the AMBER ports.
Other:
Why are there non-zero energy differences between the GROMACS ports and the AMBER 8.0 package?
There are two reasons. The first, as discussed below, is due to an inconsistency in the way AMBER orders improper torsions in TRP residues. This bug has been reported to the AMBER team, and our ports do not suffer from this issue. The second difference is seen in non-bonded interactions, which can have absolute values near zero. In these cases, the relative difference does not reflect the very low absolute error for these terms.
Why do I observe slight energy differences between this GROMACS port and the AMBER package for a system containing TRP residues?
This is due to a known bug in AMBER 8.0 involving inconsistent ordering in an improper dihedral of TRP that has been reported. This is the cause of the non-zero mean dihedral energy difference for amino acids shown in the table on the port webpage, as all TRP systems have been included in these averages.
Why are there non-zero energy differences between the GROMACS ports and the AMBER 8.0 package?
There are two reasons. The first, as discussed below, is due to an inconsistency in the way AMBER orders improper torsions in TRP residues. This bug has been reported to the AMBER team, and our ports do not suffer from this issue. The second difference is seen in non-bonded interactions, which can have absolute values near zero. In these cases, the relative difference does not reflect the very low absolute error for these terms.
Why do I observe slight energy differences between this GROMACS port and the AMBER package for a system containing TRP residues?
This is due to a known bug in AMBER 8.0 involving inconsistent ordering in an improper dihedral of TRP that has been reported. This is the cause of the non-zero mean dihedral energy difference for amino acids shown in the table on the port webpage, as all TRP systems have been included in these averages.
1250 Bellflower Boulevard, Long Beach, California 90840-9507