Protein Structure and Function. Application of Bioinformatics Methods - John Rigden 2014

Prediction of Protein Function from Surface Properties
Protein-Protein Interface
Properties of the Protein-Protein Interface

During the genomic era, remarkable progress has been achieved in understanding the number of genes within a genome and their expression. However, the products of these genes remain much less understood, especially when viewed on a global scale. While experimental Methods exist to map protein-protein interaction networks—with entire branches of bioinformatics dedicated to their analysis—The Scope of this section is limited to analyzing and predicting individual protein interfaces based on their Structure/108.html">Surface Properties. Specifically, we will focus on interactions arising from the association of Proteins that are natively stable in Water on their own. Such interactions, which form transient complexes, are known as non-obligate Protein-Protein Interactions, in contrast to obligate interactions characteristic of proteins that can only exist in an oligomeric state.

By observing A large number of interactions, we can identify statistical regularities that typically define a Protein-Protein Interface. It is worth noting that for every property mentioned here, Examples of complexes can be found where the parameter values are either significantly higher or lower than average. The interface between two Globular proteins most commonly appears as a flat, circular patch on the protein surface, the size of which is generally directly proportional to the size of the proteins (Jones and Thornton 1996). Small monomers, such as superoxide dismutase, may have an interface area of only 700 Ų, whereas monomers of the much larger tetrameric catalase feature an interface area of 10,500 Ų (Janin et al. 1988). The average interface area for a single monomer is 800 Ų, accounting for about 10% of The surface of a typical globular protein (Janin and Chothia 1990; Jones and Thornton 1995).

Typically, 55% of the interface surface is non-polar, 25% is polar, and the remaining 20% consists of charged atoms (Janin and Chothia 1990). Consequently, the average interface turns out to be less hydrophobic than the interior of the protein, yet more hydrophobic than the rest of the protein surface. Furthermore, the interface tends to be less charged than the remainder of the protein surface. To form stable complexes, interacting proteins must exhibit a match between the buried polar and charged groups of the interface. Indeed, this complementarity between such groups is widely believed to drive interaction Specificity (Chothia and Janin 1975). There is also a high degree of complementarity among hydrogen-bonding groups, with 80% of such groups on any given interface forming bonds with matching groups on the partner protein (Xu et al. 1997). On average, a protein interface will feature one hydrogen-bonding group per 80 Ų of buried surface area (Lo Conte et al. 1999). Charged groups at the interface of one protein generally do not interact with oppositely charged groups on the other; instead, they tend to be surrounded by its complementary polar groups (Lo Conte et al. 1999).