In recent years, there has been an explosive amount of protein structure information obtained and deposited in various molecular biology databases. Identifying and interpreting biologically meaningful patterns from this massive amount of information has become an essential component in directing further molecular biology research.;In this work, the SUBDUE knowledge discovering system is applied to the Brookhaven Protein Data Bank (PDB). The approaches for extracting the structure information from the PDB and for choosing suitable graph representations for these information are discussed. The results obtained from several sample data sets demonstrate the ability of SUBDUE to find biologically meaningful patterns. The limitations and future research are also discussed. |