Slideshow Cluster_Maker_2 (about 12 minutes)
Handout Cluster_Maker_2_Handout.pdf (7 pages)
Tutorial Curators Anna Kuchinsky, Scooter Morris
Data Files galFiltered.cys, collins.cys
Version Applies to clusterMaker2 0.95. Last updated: 8/29/2016
clusterMaker2 is a Cytoscape app that unifies different clustering techniques and displays into a single interface. Current clustering algorithms include Hierarchical, AutoSOME, k-medoid, k-Means, PAM, HOPACH-PAM, and DBSCAN for clustering expression or genetic data; and AutoSOME, MCL, TransClust, SPCS, Community Clustering, Affinity Propagation and MCODE for partitioning nodes in a network into similar groups.
Biological Use Case: Find possible complexes, protein families, functional relationships and view in biological context.
Dependencies: For node chart features, please also install the enhancedGraphics App.
- Start with expression data for studies into mechanism for galactose utilization. Go to File → Open and select galfiltered.cys to load a session.
Run clustering to determine interesting subnets
- Select Apps → clusterMaker → Hierarchical cluster.
- In the Node attributes for cluster box, select gal1RGexp, gal4RGexp, and gal80Rexp.
- Deselect Only use selected nodes/edges for cluster.
- Select Show TreeView when complete
- Click OK.
- You will now see TreeView visualization. On the treeview window, explore by clicking on points on the dendogram. Clicking/selecting a particular row in the heatmap will result in the expression values for that column being overlaid on the network view.
- Use shift-drag to draw a box and see results on network.
- Use shift-click to pick individual columns.
- Select an individual row by clicking on it.
- You can adjust the color scheme and contrast by going to Settings. For this demo, we'll stay with the default YellowCyan color scheme.
- Press Map Colors Onto Network and select one of the options from the Attribute List.
- Click Create Vizmap. This will map the colors onto the network.
Animate expression values over time
- Go to Map colors onto network.
- On the pop-up screen, click on specific attributes to select. For this example, select gal4RGexp and gal80Rexp.
- Press Animate Vizmap. This will animate the image on the main Cytoscape session screen.
Finding modules and complexes
Now we're going to use clusterMakers' MCL algorithm to search for modules in a protein-protein interaction network derived from TAP/MS (Tandem Affinity Purification/Mass Spectrometry). First, download the file collins.cys.
- Load collins.cys, as before (Go to File → Open and select collins.cys to load the session).
- Cluster with MCL
- Visualize clusters
- Select Apps → clusterMaker → MCL cluster to bring up the MCL cluster Settings dialog.
- The original authors of this study used a weight called PE Score to indicate the strength of the association, which we can use with MCL. Select PE Score in the Array Sources menu.
- Change the "Granularity parameter" to 1.8.
- Select "Create new clustered network" and "Restore inter-cluster edges after layout"
- Click OK. This is a large network, and may take some time to complete the clustering, depending on the speed of your machine and the number of cores. Iteration 3, in particular, will take a little longer (2-3 minutes) as it is the densest point in the cluster. For comparison, on a MacBook Pro with 16GB of RAM and a 2.66GHz Core i7, an MCL cluster of this network takes about 8 minutes.
- After the algorithm has finished, clusterMaker2 will create a new network with each of the clusters grouped together and edges between the clusters shown.
- clusterMaker2 adds a new attribute (__mclCluster in this case) to the network. Each cluster has a unique number for this attribute that may be used to change the graphics attributes in the Style panel.