Discussion of the Results
The matrix tree plot from clustering the cancer cell lines is included here as the following:
Figure 1. Clustering of the cancer cell lines according to gene expression profiles
Colon, renal, and CNS cancers, leukemias and melanomas all form fairly homogeneous clusters with these genes in this metric. Ovarian cancers show somewhat more disparity. The two prostate cancer samples show no strong association with any other group nor with each other, and the lung cancers seem to have almost no cohesion at all in this space. The breast cancers are scattered as well, two of them clustering with the melanomas, two with the CNS cancers, two beside the colon cancers, and one more in a heterogeneous cluster which also includes a prostate, two ovarian, two lung, one renal and one CNS cancer, and one melanoma cell line.
Note that 'BR:MDA-N' and 'BR:MDA-MB-435' form a sub-cluster inside the melanoma cluster. This is also indicated in Reference 1. GeneLinkerô confirms that several cancer cell lines (such as 'ME:LOX IMVI', 'RE:SN12C' and 'OV:OVCAR-8') do not cluster according to their origins, as was also found by Reference 1.
Note the similarity between the clustering of the t_matrix and the results presented in Figure 1 and Fig. 2a in Reference 1. Slight variations in the clustering parameters account for the differences.
When you are finished, you can close all the open plots either by clicking on the 'x' box in the upper-right hand corner of each, or by selecting Close All from the Window menu.
The analysis steps in this tutorial are captured in the GeneLinker Script Tutorial2.gls. If you select the top-level data table for this tutorial in the Experiments navigator and click on the Tools menu, you will see the "Run Script" item highlighted. Click on this item and select "Tutorial2.gls". The analysis steps in this tutorial will be run automatically.
GeneLinker's scripting capability is described in the GeneLinker script generation and script running documentation.
This tutorial demonstrated how to obtain and preprocess the dataset from the NCI60 studies, how to import the data, how to estimate missing values and how to do clustering calculations. A Matrix Tree Plot of the clustering of gene expression was created.
There are other commands in GeneLinkerô for handling data, analyzing data and visualizing analysis results. These are illustrated in other tutorials included in the release.
'A gene expression database for the molecular pharmacology of cancer' by Uwe Scherf, Douglas T. Ross, Mark Waltham, Lawrence H. Smith, Jae K. Lee, Lorraine Tanabe, Kurt W. Kohn, William C. Reinhold, Timothy G. Myers, Darren T. Andrews, Dominic A. Scudiero, Michael B. Eisen, Edward A. Sausville, Yves Pommier, David Botstein, Patrick O. Brown & John N. Weinstein. Nature Genetics, 24(3), pp 236-244, March 2000.
A copy of the paper can be obtained at: http://discover.nci.nih.gov/nature2000/
'Systematic variation in gene expression patterns in human cancer cell lines' by Douglas T. Ross, Uwe Scherf, Michael B. Eisen, Charles M. Perou, Christian Rees, Paul Spellman, Vishwanath Iyer, Stefanie S. Jeffrey, Matt Van de Rijn, Mark Waltham, Alexander Pergamenschikov, Jeffrey C.F. Lee, Deval Lashkari, Dari Shalon, Timothy G. Myers, John N. Weinstein, David Botstein & Patrick O. Brown, Nature Genetics, 24(3), pp 227-235, March 2000.
Where To Go From Here
Go through the other tutorials provided.
Read the online Help to learn more about the various functions of GeneLinkerô.
Further explore GeneLinkerô by using additional features.
Load up your favorite dataset and try out all the buttons and menu items.
Don't forget to right-click on things like plots - many details of graphics can be customized.
Visit the Predictive Patterns website at http://www.improvedoutcomes.com/ for the latest information on GeneLinkerô enhancements and additional products.