Application of information theory to feature selection in protein docking

Abstract

In the era of structural genomics, the prediction of protein interactions using docking algorithms is an important goal. The success of this method critically relies on the identification of good docking solutions among a vast excess of false solutions. We have adapted the concept of mutual information (MI) from information theory to achieve a fast and quantitative screening of different structural features with respect to their ability to discriminate between physiological and nonphysiological protein interfaces. The strategy includes the discretization of each structural feature into distinct value ranges to optimize its mutual information. We have selected 11 structural features and two datasets to demonstrate that the MI is dimensionless and can be directly compared for diverse structural features and between datasets of different sizes. Conversion of the MI values into a simple scoring function revealed that those features with a higher MI are actually more powerful for the identification of good docking solutions. Thus, an MI-based approach allows the rapid screening of structural features with respect to their information content and should therefore be helpful for the design of improved scoring functions in future. In addition, the concept presented here may also be adapted to related areas that require feature selection for biomolecules or organic ligands.

Prediction of ProteinProtein Interaction Pocket Using L‐Shaped PLS Approach and Its Visualizations by Generative Topographic Mapping

“…[22] Related to this theme, host-guest relationships other than GTM pocket mapping could be deduced from informatics technology such as the mutual information. [23,24] Based on conditional probability theory, mutual information identifies the most likely interactions between the amino acid residues in the host and guest protein, according to number of contacts. We are currently perusing these avenues to improve our predictive tools for PPI pockets.…”

“…For a comprehensive design, rather than physico‐chemical properties such as MLP, medicinal chemistry descriptors such as pseudo‐centers should be considered 22. Related to this theme, host‐guest relationships other than GTM pocket mapping could be deduced from informatics technology such as the mutual information 23,24. Based on conditional probability theory, mutual information identifies the most likely interactions between the amino acid residues in the host and guest protein, according to number of contacts.…”

Scoring docking conformations using predicted protein interfaces

“…Experiments revealed that both approaches can distinguish between native and non-native structures with accuracy around 80%. More recently, Othersen et al [68] conducted a similar experiment using mutual information to select discriminative structural features [46]. They identified 11 of them which led to good identification of near-native models.…”

Efficacious End User Measures—Part 1: Relative Class Size and End User Problem Domains

“…It was subsequently included in a measure comparison by Baldi et al [21]. Since then, it has gained some traction in biological literature [35][36][37][38][39][40][41][42][43] and has been seen in network management literature [44]. The measure is sometimes called the information coefficient or mutual information coefficient; we use the acronym IC.…”

“…Information-theory-based measures are gaining traction in the literature [35][36][37][38][39][40][41][42][43]. Some of these reports indicate the belief that the measures are rCS-invariant [38,40,43]. Solis and Rackovsky [41] note that their particular information theoretic measure may not be rCS-invariant.…”