Unit 6: Metabolism and Networks
Project: Software for Analyzing Complex Data Sets

Description of the project.
Reading a paper.
Writing up your analysis.

Objectives:

• Understand cluster analysis and tree building as methods for making sense of complex data.
• Discuss the algorithms in CLUSTER and TREEVIEW and their applications to metabolic pathways.

 
Project Description:

The focus of this project is a paper by MB Eisen et al., entitled "Cluster analysis and display of genome-wide expression patterns." Click here for a copy.

The project consists of a written analysis of the cluster analysis paper. Your analysis should be no more than five typed, double-spaced pages. Try to keep it simple and clear. Total points = 20. Report due at the beginning of class on Thursday December 11, 2003.

The paper selected for this project addresses the bioinformatic problem of analyzing the status of cellular processes based on genomic expression patterns.

Two programs, CLUSTER and TREEVIEW have been developed based on the algorithm described in the paper. A manual is available for use with the software.

Reading the paper:

1. As you make a first pass through the paper, make a list of any terms or concepts to look up later.

2. Create an rough outline or idea map of the main points of the paper (essential a very brief overview).

3. Look up the terms and concepts that relate to the main point of the paper.

4. Use this information to fill in the broad outline or idea map you put together in step 2 above.

5. Try to determine what the authors were attempting to accomplish by doing the work described in the paper. Use the resources you have available, including citations in the paper itself, articles in the course library, resources in the exercises, and glossaries.

The Report:

The following issues should be addressed in your report.

1. What problem were the authors trying to solve?

2. How did the authors approach the problem?

3. Did the authors try more than one way to solve the problem? If so, what worked and what didn't?

4. Explain in terms you understand the basic idea of cluster analysis.

5. How did the authors test their approach?

6. How can bioinformaticists or molecular biologists use the software that the authors created?