This is a fork of the original StratomeX repository to integrate dynamic, server-side clustering on genomic data and analyze the confidence or ambiguity of cluster assignments. It is implemented as Caleydo web plugin: https://github.com/Caleydo/caleydo_web
Demo-Version of the tool: http://ec2-52-58-105-31.eu-central-1.compute.amazonaws.com/
- Supports application of clustering algorithms (k-means, hierarchical, affinity propagation, fuzzy) to genomic data.
- Creates views showing the distances/correlation/probability of each record in one cluster to the cluster centroid and the distances/correlation/probability of each patient to all other cluster.
- Provides splitting of one cluster into three different subsets (well-fitting, uncertain, bad-fitting) to further analyze how well the records fit into one cluster.
- Supports refinement of clusters/groups (shifting to better fits, merging, exclusion)
- Support dynamical increase/reduction of the number of clusters in hierarchical cluster results.
- Checkout caleydo_web_container: https://github.com/Caleydo/caleydo_web_container
- Checkout calyedo_clustering: https://github.com/K3rn1n4tor/caleydo_clustering
- Update all other repos to the latest version (at least from March 21st).
- Make sure you have the caleydo_data_hdf (required for TCGA data) and caleydo_vis_lineup plugin.
- See Issues section for further information.