Package utility

Class Clustering

java.lang.Object

utility.Clustering

public final class Clustering
extends Object

Provides functionality for clustering data using the HDBSCAN algorithm (see https://tribuo.org/).

The static implementation allows to re-use the trainer and model and provides methods to add samples (not Sample instances, but data points), training and resetting the model, as well as retrieving clustering results.

The implementation currently only supports the clustering based on variants, i.e. each feature is defined by a position and alternative base in the context of Feature.Alleles, Feature.Proteoforms or Samples and considered a binary feature (weight or value of 1.0). These are compared with a Manhattan distanc (L1 distance).

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static final record	Clustering.Entry	A record to represent a clustering result entry.

Constructor Summary

Constructors

Constructor	Description
Clustering()

Method Summary

Modifier and Type	Method	Description
static void	addToDataset(String label, Map<Integer,String> variants)	Adds a sample to the dataset for clustering.
static List<Clustering.Entry>	getClusteringResult()	Retrieves the clustering results as a list of Clustering.Entry instances.
static boolean	hasData()	Checks if the dataset contains any samples.
static void	reset()	Resets the clustering state by clearing the dataset, labels, and model.
static void	train()	Trains the HDBSCAN model using the current dataset.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

Clustering

public Clustering()

Method Details

reset

public static void reset()

Resets the clustering state by clearing the dataset, labels, and model.

addToDataset

public static void addToDataset(String label,
 Map<Integer,String> variants)

Adds a sample to the dataset for clustering.

Parameters:

label - The label of the sample.

variants - A map of feature positions and their corresponding values.

hasData

public static boolean hasData()

Checks if the dataset contains any samples.

Returns:

true if the dataset has at least one sample, false otherwise.

train

public static void train()

Trains the HDBSCAN model using the current dataset.

getClusteringResult

public static List<Clustering.Entry> getClusteringResult()

Retrieves the clustering results as a list of Clustering.Entry instances.

Returns:

A list of clustering result entries, each containing the name, label, index, and outlier score.