Uni-Logo

Institut für Informatik
 

Technical Report No. 175 - Abstract


Kristian Kersting, Tapani Raiko, Stefan Kramer, Luc De Raedt
Towards Discovering Structural Signatures of Protein Folds based on Logical Hidden Markov Models

With the growing number of determined protein structures and the availability of classification schemes, it becomes increasingly important to develop computer methods that automatically extract structural signatures for classes of proteins. In this paper, we apply a new Machine Learning technique, Logical Hidden Markov Models (LOHMMs), to the task of finding structural signatures of folds according to the classification scheme SCOP. Our preliminary results indicate that LOHHMs are in fact applicable to this task: The number of parameters of our LOHMM is by an order of magnitude smaller than the number of an equivalent HMM, and the generalization performance appears satisfactory. Furthermore, we show that it is easy to extract characteristic patterns from the trained LOHMM. Summing up, we believe that LOHMMs show a great promise for the analysis of such data and for the discovery of structural signatures of protein folds.


Report No. 175 (PostScript)