Geometrical data mining with applications in bioinformatics.

3D data is ubiquitous. The amount of real-world data such as protein structures, surveillance videos or geographical information has recently exploded. At the same time, data mining methods have tried to extract useful knowledge from these data. However, the current methods still have shortcomings. For example, when predicting the binding of a drug molecule to a protein, it is insufficient to consider only the sequence of the protein. In this setting, various geometrical and chemical constraints determine whether the molecule will bind to the protein, which cannot be discovered from the sequence alone. Current data mining methods do not allow to search for patterns including all these constraints. In this project, we will develop a novel data mining framework based on geometrical patterns. Such patterns consist of geometrical (e.g., 3D coordinates and angles) as well as application-specific constraints (e.g., chemical knowledge). With this framework, we will be able to efficiently discover patterns in a completely new range of real-world data with an expressiveness that represents a significant advance over the current state-of-the-art technology. Such methods could not only lead to new insights in the interactions of proteins, but are also more generally applicable in fields such as robotics and computer vision.

Keywords:Bioinformatics, Geometrical pattern mining, Data mining, Artificial intelligence

Disciplines:Applied mathematics in specific fields, Computer architecture and networks, Distributed computing, Information sciences, Information systems, Programming languages, Scientific computing, Theoretical computer science, Visual computing, Other information and computing sciences, Artificial intelligence, Cognitive science and intelligent systems, Biomaterials engineering, Biological system engineering, Biomechanical engineering, Other (bio)medical engineering, Environmental engineering and biotechnology, Industrial biotechnology, Other biotechnology, bio-engineering and biosystem engineering