Package chemaxon.clustering

Class Ward

  • @PublicAPI
public class Ward
extends Common
    Uses Ward's minimum variance method for clustering molecules based on molecular fingerprints or other descriptors.
    This class can be used from command-line or from the API.
    See the documentation for details.

    An example for running Ward: 

     ConnectionHandler ch;
 String selectString;
 ...
 Ward ward=new Ward();
 ward.setInput(ch, selectString);
 ward.setOutput(System.out);
 ward.setDimensions(0);
 ward.setFpSize(512);
 ward.setClusterCount(10);
 ward.setCentralShown(false);
 ward.setSingletonNegative(false);
 ward.setStatNeeded(true);
 ward.setOnlyStat(false);
 ward.setStatStream(System.out);
 ward.setKelleyStats("kelley_results.txt");
 ward.setMode(Ward.RNN_AND_CLUSTERING);       // execution mode
 ward.setIdGeneration(true);
 ward.run();

    Since:
    JChem 1.6.2

    • Constructor Detail

      • Ward

        public Ward()

    • Method Detail

      • setInput

        public void setInput​(ConnectionHandler conh,
                     String querySQL)
              throws SQLException
        Sets input as an SQL query.
        Parameters:
        conh - ConnectionHandler object opened to the database
        querySQL - an SQL SELECT statement used as data source
        Throws:
        SQLException
        Since:
        JChem 1.6.2

      • setInput

        public void setInput​(InputStream is)
        Sets input as an InputStream
        Parameters:
        is - the InputStream to be used for input
        Since:
        JChem 1.6.2

      • setCentralShown

        public void setCentralShown​(boolean b)
        Specifies if cluster centers should be diplayed.
        Parameters:
        b - if set to true, central objects are calculated and shown in output.
        Since:
        JChem 1.6.2

      • getCentralShown

        public boolean getCentralShown()
        Indicates whether central objects are displayed.
        Returns:
        true if central objects are calculated and shown in output.
        Since:
        JChem 1.6.2

      • setKelleyStats

        public void setKelleyStats​(PrintStream ps)
        Directs Kelley statistics to a PrintStream.
        Parameters:
        ps - the PrintStream where Kelley statistics directed to, or null,if no Kelley statistics needed.
        Since:
        JChem 1.6.2

      • setKelleyStats

        public void setKelleyStats​(File file)
                    throws FileNotFoundException
        Directs Kelley statistics to a File.
        Parameters:
        file - the File where Kelley statistics directed to, or null,if no Kelley statistics needed.
        Throws:
        FileNotFoundException
        Since:
        JChem 1.6.2

      • setKelleyStats

        public void setKelleyStats​(String fileName)
                    throws FileNotFoundException
        Directs Kelley statistics to a File specified with a name.
        Parameters:
        fileName - the path of the file where Kelley statistics directed to, or null,if no Kelley statistics needed.
        Throws:
        FileNotFoundException
        Since:
        JChem 1.6.2

      • setSingletonNegative

        public void setSingletonNegative​(boolean b)
        Specifies whether singletons get negative cluste id-s.
        Parameters:
        b - If set to true, singeltons get negative cluster id-s.
        Since:
        JChem 1.6.2

      • isSingletonNegative

        public boolean isSingletonNegative()
        Indicates whether singeltons get negative cluster id-s.
        Returns:
        true, if singeltons get negative cluster id-s.
        Since:
        JChem 1.6.2

      • setClusterCount

        public void setClusterCount​(int count)
        Sets the number of clusters.
        Parameters:
        count - the number of clusters
        Since:
        JChem 1.6.2

      • getClusterCount

        public int getClusterCount()
        Gets the number of clusters.
        Returns:
        the number of clusters
        Since:
        JChem 1.6.2

      • setMode

        public void setMode​(int mode)
        Sets the clustering mode.
        Parameters:
        mode - clustering mode constant
        Selectable modes:
        • RNN_AND_CLUSTERING
        • ONLY_RNN
        • ONLY_CLUSTERING
        Since:
        JChem 1.6.2

      • getMode

        public int getMode()
        Gets the clustering mode.
        Returns:
        one of the folloing constants:
        • RNN_AND_CLUSTERING
        • ONLY_RNN
        • ONLY_CLUSTERING
        Since:
        JChem 1.6.2

      • run

        public void run()
         throws chemaxon.clustering.ClusteringException,
                IOException,
                SQLException,
                chemaxon.clustering.InvalidLicenseKeyException
        Starts processing.
        Throws:
        chemaxon.clustering.ClusteringException
        IOException
        SQLException
        chemaxon.clustering.InvalidLicenseKeyException
        Since:
        JChem 1.6.2

      • main

        public static void main​(String[] args)
        The command line version entry point. Parameter parsing, initialization then start.