Related Experiment Videos
Generalized singular value decomposition for comparative analysis of genome-scale expression data sets of two
Orly Alter1, Patrick O Brown, David Botstein
1Department of Genetics, Stanford University, Stanford, CA 94305. orly@genome.stanford.edu
Summary
This study introduces a mathematical framework to compare genome-scale expression data. It reconstructs and classifies genes and arrays by analyzing regulatory programs and biological processes across datasets.
Area of Science:
- Genomics
- Bioinformatics
- Computational Biology
Background:
- Genome-scale expression data provides insights into cellular functions.
- Comparing datasets from different species or conditions is challenging due to variations.
- Existing methods may not adequately account for shared and unique biological signals.
Purpose of the Study:
- To develop a unified mathematical framework for comparing two genome-scale expression datasets.
- To formulate gene expression as a superposition of biological and technical factors.
- To enable comparative reconstruction and classification of genes and arrays.
Main Methods:
- Utilized generalized singular value decomposition (GSVD).
- Modeled expression data as a combination of common and unique regulatory programs, biological processes, and experimental artifacts.
- Applied the framework to compare yeast and human cell-cycle expression datasets.
Main Results:
- The framework successfully separated common and dataset-specific influences on gene expression.
- Enabled comparative reconstruction and classification of genes and arrays across species.
- Demonstrated the framework's utility in analyzing cell-cycle regulation in yeast and humans.
Conclusions:
- The proposed mathematical framework offers a robust approach for comparative analysis of genome-scale expression data.
- This method enhances the understanding of conserved and divergent biological processes across species.
- Facilitates more accurate gene and array classification in multi-dataset studies.