Related Experiment Videos
Cluster analysis and display of genome-wide expression patterns
M B Eisen1, P T Spellman, P O Brown
1Department of Genetics, Stanford University School of Medicine, 300 Pasteur Avenue, Stanford, CA 94305, USA.
Summary
This study introduces a cluster analysis system for gene expression data, grouping genes by expression patterns. This method aids in identifying gene functions and understanding cellular processes from genome-wide experiments.
Area of Science:
- Genomics
- Bioinformatics
- Systems Biology
Background:
- DNA microarrays generate vast amounts of genome-wide gene expression data.
- Interpreting these complex datasets to understand gene function and cellular processes is challenging.
- Standard statistical methods are needed to analyze and visualize gene expression patterns.
Purpose of the Study:
- To develop a cluster analysis system for genome-wide gene expression data.
- To visualize gene clustering and expression patterns intuitively for biologists.
- To aid in the functional annotation of genes, especially novel or poorly characterized ones.
Main Methods:
- Utilized standard statistical algorithms for cluster analysis.
- Applied the system to genome-wide expression data from DNA microarray hybridizations.
- Developed a graphical output to display clustering and expression data simultaneously.
Main Results:
- The cluster analysis system successfully grouped genes with similar expression patterns.
- In budding yeast (Saccharomyces cerevisiae), genes of known similar function were efficiently clustered together.
- A similar tendency was observed in human gene expression data.
- Coexpression patterns provided insights into the functions of novel genes.
Conclusions:
- Genome-wide expression patterns can indicate the status of cellular processes.
- This clustering approach offers a straightforward method for inferring the functions of uncharacterized genes.
- The system provides a valuable tool for biologists analyzing complex genomic data.