Orientations of Proteins in Membranes (OPM) database provides spatial positions of protein three-dimensional structures with respect to the lipid bilayer. The database was used in experimental and theoretical studies of membrane-associated proteins
Proteins structures are taken from the Protein Data Bank
. Positions of the proteins in a hydrophobic slab are calculated using the implicit solvation model
OPM provides structural classification of membrane-associated proteins into families and superfamilies (based on SCOP classification), membrane topology, and the type of a destination membrane for each protein. All protein coordinate files with calculated membrane boundaries are freely downloadable.
The site allows visualization of protein structures with membrane boundary planes through Jmol, MDL Chime and WebMol.
This database provides spatial positions of transmembreane
and peripheral membrane proteins
in the lipid bilayer, along with PDBTM database
that includes only transmembrane proteins. Calculated positions of proteins have been compared with relevant experimental data for 24 tranmembrane and 53 peripheral membrane proteins including site-directed spin labeling
, chemical labeling, measurement of membrane binding affinities of protein mutants
spectroscopy, solution or solid-state NMR spectroscopy
ATR FTIR spectroscopy
, and X-ray or neutron diffraction studies of proteins in lipid bilayers
The set of peripheral membrane proteins
in OPM is incomplete. The database probably includes less than 50% of peripheral proteins from the Protein Data Bank
, because membrane-anchoring elements of peripheral proteins
(amphiphilic alpha helices
, exposed nonpolar residues, or lipid anchors
) are missing or disordered in the experimental protein structures, and therefore the mode of protein-membrane association can not be computationally predicted.
Another feature is the use of only one "representative" PDB entry for every protein complex, unlike PDBTM that includes all crystal structures of the same protein in the PDB. The "representative" structure of each protein was selected as one with the "most complete coverage" (smallest total length of disordered segments in the crystal structure, or a quarternaly complex with the larger number of subunits), rather than a structure determined with the highest resolution.
Other related resources