Residues template for lipids about openmm HOT 6 CLOSED

Amber doesn't support POPI. Here are all the lipid types it has.

$ fgrep "<Residue " lipid17.xml 
    <Residue name="DLPC">
    <Residue name="DLPE">
    <Residue name="DLPS">
    <Residue name="DLPA">
    <Residue name="DLPG">
    <Residue name="DMPC">
    <Residue name="DMPE">
    <Residue name="DMPS">
    <Residue name="DMPA">
    <Residue name="DMPG">
    <Residue name="DPPC">
    <Residue name="DPPE">
    <Residue name="DPPS">
    <Residue name="DPPA">
    <Residue name="DPPG">
    <Residue name="DSPC">
    <Residue name="DSPE">
    <Residue name="DSPS">
    <Residue name="DSPA">
    <Residue name="DSPG">
    <Residue name="DOPC">
    <Residue name="DOPE">
    <Residue name="DOPS">
    <Residue name="DOPA">
    <Residue name="DOPG">
    <Residue name="POPC">
    <Residue name="POPE">
    <Residue name="POPS">
    <Residue name="POPA">
    <Residue name="POPG">
    <Residue name="SDPC">
    <Residue name="DAPC">
    <Residue name="SDPE">
    <Residue name="DAPE">
    <Residue name="SDPS">
    <Residue name="DAPS">
    <Residue name="SDPA">
    <Residue name="DAPA">
    <Residue name="SDPG">
    <Residue name="DAPG">
    <Residue name="CHL1">

from openmm.

Hi @peastman thanks for the reply.

Unfortunately, the same error is raised even if I use lipids supported by Amber.
The same code as above raised this error: ValueError: No template found for residue 291 (CHL1). The set of atoms matches CHL1, but the bonds are different. Perhaps the chain is missing a terminal group? For more information, see https://github.com/openmm/openmm/wiki/Frequently-Asked-Questions#template.

Attached the PDB file.
step5_input_openmm.tar.gz

from openmm.

The error message is different this time. Notice that it says, "The set of atoms matches CHL1, but the bonds are different." Your PDB file is missing CONECT records for the lipids. For anything other than standard amino acids, nucleic acids, and water, a PDB file needs to specify which atoms are bonded to each other. This is discussed in the FAQ entry linked in the error message.

from openmm.

That makes sense, thank you.

One more question, is there a way to identify the "boundaries" of a membrane after calling the addMembrane method?
Ideally, I would like to know what are the boundaries of the bilayer (excluding the protein), but I couldn't find a way to do so.

Thanks!

from openmm.

How do you define the boundaries? Given the Topology, it's easy to get a list of atoms that make up the membrane.

membrane = [atom for atom in topology.atoms() if atom.residue.name == 'CHL1']

You could then find the minimum and maximum z coordinates of any lipid atom.

lowerz = min(positions[atom.index][2] for atom in membrane)
upperz = max(positions[atom.index][2] for atom in membrane)

But the membrane isn't smooth, so a single z value is only a very rough measure of its extent. Or perhaps you were asking about the boundary where it contacts the protein, not where it contacts the water? There's also the issue of periodic boundary conditions. If the membrane has gotten wrapped in a way that splits it between two periodic copies, the "upper" layer could be at the bottom of the box and the "lower" layer at the top of the box.

from openmm.

Thank you, that worked!

from openmm.