Membrane topology

Topology of a Transmembrane protein refers to the exact number and location of membrane spanning segments and their orientation relative to the membrane. Determining the atomic level level structure of transmembrane proteins is rather challenging due the experimental conditions, however in most cases their topology provides sufficient information.

There are several computational methods for predicting alpha-helical transmembrane domains (i.e. topography). Pioneer methods utilizied the fact that membrane spanning regions contain more hydrophobic residues than other parts of the protein, however applying different hydrophobic scales altered the prediction results. Later several statistical method was developed to improve the topography prediction, as well as a special alignment method was introduced. According to the positive-inside rule, cytoslic loops near the lipid bilayer contain more positively charged amino acids. Applying this information resulted in the first topology prediction methods. As more structures were determined machine learning algorithms appeared. Supervised learning methods are trained on a set of experminetally determined structures, however these methods highly depend on the training set used. Unsupervised learning methods are based on the principle that topology depends on the maximum divergence of the amino acid distributions in different structural parts. It was also shown, that locking a segment location based on prior knowledge about the structure improves the prediction accuracy. This feature has been added to some of the existing prediction methods. It is important to note that several methods are not able to distinguish Signal peptide-s at the N-terminus of the protein and transmembrane segments, or discriminate Transmembrane protein-s and Globular protein-s.

The most recent methods use consensus prediction (i.e. they use several algorithm to determined the final topology) and automatically incorporate previously determined experimental informations.

It is also possible to predict beta-barrel membrane proteins' topology.

There are a few databases which provide topologies of membrane proteins: Uniprot is the most comprehensive protein database, the recently updated TOPDB and ExTopoDB contains experimentally determined localizations, TOPDOM is a database of domains located conservatively in one side of transmembrane proteins, the HTP database is the collection of human transmembrane proteins.