Below are a number of coordinate, forcefield and topology files that can be freely downloaded and used. The coordinate files are in .pdb format and can be visualized with most molecular graphics programs. The forcefield and topology files are probably only useful together with the Gromacs set of molecular dynamics programs. Most of these files come from published work, but if you are going to use any of them in your own work, do double and triple check them. I do not guarantee that they are error-free, and if they cause your computer to burst into flames and your lab to burn down, don't blame me. If you find any errors, please let me know. If you publish work based on these structures, please cite (one of) the corresponding references (numbers refer to numbers in my publication list). If you download files, I'd be most interested to hear what you plan to do with them. That's all :)

In case anyone wonders what good alamethicin files are: for many smaller membrane proteins they make useful starting structures. Just replace the alamethicin bundle of approximately the same size by your pet membrane protein and go from there.

I'm interested in hearing about other sites that have phospholipid coordinates. Two I know of:


Micelles
m40.pdbMicelle consisting of 40 DPC lipids and 3880 water molecules, after 1100 ps. See ref. 21.
m54.pdbMicelle consisting of 54 DPC lipids and 5238 water molecules, after 1100 ps. See ref. 21.
m65.pdbMicelle consisting of 65 DPC lipids and 6305 water molecules, after 1100 ps. See ref. 21.
Lipid bilayers
dppc64.pdb64 DPPC lipids and 3864 water molecules. Starting structure for most runs in ref. 6.
dppc128.pdb128 DPPC lipids and 3655 water molecules. Final structure of run E in ref. 1.
dppc128_npat.pdb128 DPPC lipids and 3655 water molecules. 1 ns of constant area at 0.629 nm^2.
dmpc_npat.pdb128 DMPC lipids and 3655 water molecules. 1 ns of constant area at 0.596 nm^2.
popc128a.pdb128 POPC lipids and 2460 water molecules. Starting structure for refs. 13, 14, 15 and other POPC/membrane protein publications. Simulated for 1600 ps without membrane protein before making holes in it and such.
popc128b.pdb128 POPC lipids and 2460 water molecules. Started from popc128a.pdb, simulated for another 1 ns at 300K with anisotropic pressure coupling.
pope.pdbLarge POPE bilayer with 340 lipids. Based on ref. 10.
Alamethicin helix bundles
almN4start.pdb4-helix alamethicin bundle starting structure. See refs. 17, 23.
almN4end.pdb4-helix alamethicin bundle structure after 4 ns. See ref. 23.
almN5start.pdb5-helix alamethicin bundle starting structure. See ref. 12, 17, 23.
almN5end.pdb5-helix alamethicin bundle structure after 4 ns. See ref. 12, 17, 23.
almN6start.pdb6-helix alamethicin bundle starting structure. See ref. 12, 14, 17, 23.
almN6end.pdb6-helix alamethicin bundle structure after 4 ns. See ref. 12, 14, 17, 23.
almN7start.pdb7-helix alamethicin bundle starting structure. See ref. 12, 17, 23.
almN7end.pdb7-helix alamethicin bundle structure after 4 ns. See ref. 12, 17, 23.
almN8start.pdb8-helix alamethicin bundle starting structure. See ref. 12, 17, 23.
almN8end.pdb8-helix alamethicin bundle structure after 4 ns. See ref. 12, 17, 23.
Gromacs forcefield and topology files
dpc.itpTopology file for a dodecylphosphocholine (DPC) lipid.
popc.itpTopology file for a POPC molecule. Needs lipid.itp.
dppc.itpTopology file for a DPPC molecule. Needs lipid.itp.
dmpc.itpTopology file for a DMPC molecule. Needs lipid.itp.
dlpe.itpTopology file for a DLPE molecule. Not yet available.
pope.itpTopology file for a POPE molecule. Needs lipid.itp.
lipid.itpLipid parameters plus the interaction parameters between the lipid atoms and the modified GROMOS87 forcefield in Gromacs.
g96_lipid.itpLipid parameters plus the interaction parameters between the lipid atoms and the GROMOS96 forcefield. GROMOS96 is available with Gromacs, this file not yet available.
example.topExample topology file for a lipid/protein/water/ions system.
example2.topExample topology file for a pure DPPC bilayer with water.

NOTE: 3 carbons in dppc.itp had the wrong mass (off by 1 hydrogen mass) and 1 bond was too short (double instead of single bond). Fixed, July 26 2000

NOTE: names in pope.pdb suggested they were popc lipids. This only affects the names, the lipids really were pope. I changed the names for atoms 1-3 to H1, H2, H3. They are now consistent with pope.itp. I also removed 3 pairs in pope.itp that in Gromacs were zero anyway. March 1 2001


last modified: 20 March 2001