Step 1
Upload data or select from sample data

Upload a CSV/TXT file containing numeric data:

Add file

Cluster:  Rows   Columns

Is there a row name/ID column in your data?  Yes   No (IDs must be unique)

  • File size limit is currently 2 GB but expected to increase in the future.
  • Uploaded file will be deleted automatically after 7 days.
  • You can drag & drop a file from your desktop on this webpage (see Browser support).
  • Built on KnowEnG platform and icons from Glyphicons.
Sample data
Select a sample database and choose from the clustering methods below.
NCI60 expression data
B-cell lymphoma gene expression (by columns only)
  • Clustering rows may take time to render visualization.
  • Query GEO database
    Perform hierarchical clustering and draw a heat map
    Enter GEO (Gene Expression Omnibus) DataSet ID (format GDS####):
    Step 2
    Choose clustering methods
    k-means      Parallel k-means
    Affinity propagation#
    Spectral clustering      Parallel spectral clustering
    Gaussian mixture model
    Hierarchical clustering
    DBSCAN#  eps:
    Number of clusters:

    Please refresh your browser and clear the cache if you encounter a visualization problem.
    = More information
    #Number of clusters optional for AP and DBSCAN.
    *The user is encouraged to vary epsilon value depending on input dataset.

    Select algorithm to get additional options here.

    Algorithm description
    Step 3
    Visualize results