Hi all, As I was planning to test some different clustering algorithms anyway in the future of the segal project, I copied the Java code of MCL into .../java/MCL I should also say that with Thomas and our thesis student Andreas, we are implementing a JAVA clustering library, containing different types of clustering algorithms, which will be ready to use for different problems in bioinformatics and machine learning. This will make it easier in the future to try different clustering algorithms, e.g. prior to the network inference step. In the framework of my BOF-project I am now doing a literature study on different theoretical aspects of clustering, hoping to result in an updated taxonomy of different clustering algorithms, their internal relations, respective (dis)advantages and characteristics of the data/clusters they assume. In particular, my attention focuses on the aspects of data sets for clustering as they arise in bioinformatics, i.e. large dimensionality, noise, outliers, and other funny features that hampers the use of traditional clustering algorithms, and for which thus algorithms need to be extended/adapted. greetz, yvan -- ================================================================== Dr. Yvan Saeys, PhD DEPARTMENT OF PLANT SYSTEMS BIOLOGY Fax:32 (0)9 331 38 09 BIOINFORMATICS TEAM Tel:32 (0)9 331 36 95 GHENT UNIVERSITY, Technologiepark 927, B-9052 Gent, Belgium Vlaams Instituut voor Biotechnologie VIB mailto:yvan.saeys@ugent.be http://bioinformatics.psb.ugent.be/ http://www.psb.ugent.be/~yvsae/ ================================================================== -- ================================================================== Dr. Yvan Saeys, PhD DEPARTMENT OF PLANT SYSTEMS BIOLOGY Fax:32 (0)9 331 38 09 BIOINFORMATICS TEAM Tel:32 (0)9 331 36 95 GHENT UNIVERSITY, Technologiepark 927, B-9052 Gent, Belgium Vlaams Instituut voor Biotechnologie VIB mailto:yvan.saeys@ugent.be http://bioinformatics.psb.ugent.be/ http://www.psb.ugent.be/~yvsae/ ==================================================================