A constraint programming approach to bioinformatics structural problems

In this paper we show how Constraint Programming (CP) techniques have been used to handle bioinformatics structural problems, namely in protein structure prediction and protein interaction (docking). Solving these problems requires innovative modelling of the problem variables and constraints, and the application of advanced CP features to handle the problems efficiently, namely the exploitation of global constraints and local search, in addition to more standard binary constraint propagation. Both applications, respectively PSICO (Processing Structural Information with Constraint programming and Optimisation), and BiGGER (Bimolecular complex Generation with Global Evaluation and Ranking) have been incorporated in a platform, Chemera, that aims at supporting (and has effectively supported, namely in protein docking), biochemists in their research.