Automated identification of features of protein-ligand interactions using inductive logic programming: application to hexose binding

Motivation: There is an urgent need for automated methods to learn general features of the interactions of a ligand class with its diverse set of protein receptors. An appropriate machine learning approach is Inductive Logic Programming (ILP) which automatically generates comprehensible rules in addition to prediction. The development of ILP systems which can learn rules of the complexity required for studies on protein structure remains a challenge. Here we use a new ILP system, ProGolem, and demonstrate its performance on learning features of hexose/protein interactions. Results: The rules induced by ProGolem detect aromatic and planarpolar residues interactions in addition to less common features like the aromatic sandwich. The rules also reveal a previously unreported dependency over residues CYS and LEU. They also specify stereo configurations involving aromatic and hydrogen bonding residues. Further to confirming literature results, ProGolem’s model has a 10fold cross-validated predictive accuracy that is superior, at the 95% confidence level, to other ILP systems and statistical learners. Availability: http://ilp.doc.ic.ac.uk/Hexose Contact: [email protected]