Clustering Overview

Overview

Clustering is a type of multivariate statistical analysis also known as cluster analysis, unsupervised classification analysis, or numerical taxonomy. In molecular biology, clustering is used to group biological samples or genes into separate clusters based on their statistical behavior. The main objective of clustering is to find similarities between experiments or genes (given their expression ratios across all genes or samples, respectively), and then group similar samples or genes together to assist in understanding relationships that might exist among them.

Cluster analysis is based on a mathematical formulation of a measure of similarity. There are a number of characteristics that distinguish different approaches to cluster analysis.

Cluster Analysis Characteristics:

• Numerical, statistical, and conceptual clustering.

• Agglomerative vs. divisive.

• Overlapping vs. disjoint clusters.

• Incremental vs. non-incremental.

• Flat vs. hierarchical representations.

In GeneLinker™, the following clustering methods are available:

• K-Means

• Jarvis-Patrick

• Agglomerative Hierarchical

• Self Organizing Maps

All of the above methods are applicable to both genes and samples.

Related Topic:

Distance Metrics Overview