Release of clusterMaker 1.10

[Scooter Morris]

The RBVI is pleased to announce the release of clusterMaker 1.10.  This version of clusterMaker includes several new clustering algorithms, including k-medoid and the addition of an iterative approach to finding the best k for k-medoid and k-means using silhouette.  We've also added the ability to create correlation networks from a set of node attributes.  This can be used, for example, to create expression correlation networks from expression data.  With these, clusterMaker now provides 12 cluster algorithms, three poster-cluster filters, and several visualizations which tie cluster results into the network, including the ability to create nodeCharts if the nodeCharts plugin is installed.

In addition, clusterMaker, now exports a variety of CyCommands, which can be used from scripts or from other plugins.  clusterMaker documentation is available at http://www.cgl.ucsf.edu/cytoscape/cluster/clusterMaker.html and a manuscript describing three clusterMaker scenarios has just been published in BMC Bioinformatics: clusterMaker: a multi-algorithm clustering plugin for Cytoscape.  Morris, JH et al. BMC Bioinformatics 2011, 12:436 (9 November 2011).  Tutorials describing those scenarios are also available at http://opentutorials.cgl.ucsf.edu/index.php/Portal:Cytoscape.

clusterMaker 1.10 is available for download from the Plugin Manager under the Analysis category.  Please cite the BMC Bioinformatics paper above if you use clusterMaker.

clusterMaker attribute algorithms:
    AutoSOME Clustering*
    Create Correlation Network from Node Attributes
    Hierarchical cluster
    K-Means cluster**
    K-Medoid cluster**

clusterMaker network partition algorithms:
    Affinity Propagation cluster
    AutoSOME Clustering*
    Connected Components cluster
    Community cluster (GLay)
    MCODE cluster
    MCL cluster*
    SCPS cluster
    Transitivity Clustering*

clusterMaker filters:
    Cutting Edge Filter
    Density Filter
    HairCut Filter

clusterMaker visualizations:
    Create New Network from Attribute
    Create New Network from Clusters
    Create New Network with nested networks from attribute
    Eisen KnnView
    Eisen TreeView
    HeatMapView (unclustered)

* Multi-threaded
** Can use silhouette to estimate k.  Silhouette estimation is multi-threaded.