Clustering / K-means

Description

Performs a K-means clustering for genes. In K-means clustering genes are put into a user specified number (K) of clusters. Please note that in order to estimate the K, you can run the tool "K-means - estimate K".

Parameters

Details

This clustering is done using Euclidean distance, and cannot be changed.

Output

A data file where the cluster information is added to the last column. You can visualise the clusters by selecting the "Clustered profiles" visualisation in the visualisation panel. If you would like to extract genes for a particular cluster, please run the tool "Extract genes from clustering" which can be found in the "Utilities" category. In addition to the data file, K-means clustering returns also a cluster image generated by R.