Abstract
A central goal of postgenomic biology is the elucidation of the regulatory relationships among all cellular constituents that together comprise the 'genetic network' of a cell or microorganism. Experimental manipulation of gene activity coupled with the assessment of perturbed transcriptome (i.e., global mRNA expression) patterns represents one approach toward this goal, and may provide a backbone into which other measurements can be later integrated. We use microarray data on 287 single gene deletion Saccharomyces cerevisiae mutant strains to elucidate generic relationships among perturbed transcriptomes. Their comparison with a method that preferentially recognizes distinct expression subpatterns allows us to pair those transcriptomes that share localized similarities. Analyses of the resulting transcriptome similarity network identify a continuum hierarchy among the deleted genes, and in the frequency of local similarities that establishes the links among their reorganized transcriptomes. We also find a combinatorial utilization of shared expression subpatterns within individual links, with increasing quantitative similarity among those that connect transcriptome states induced by the deletion of functionally related gene products. This suggests a distinct hierarchical and combinatorial organization of the S. cerevisiae transcriptional activity, and may represent a pattern that is generic to the transcriptional organization of all eukaryotic organisms. Color versions of both the Supplementary Material and the article are available at http://angel.elte.hu/bioinf.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 601-612 |
| Number of pages | 12 |
| Journal | Physica A: Statistical Mechanics and its Applications |
| Volume | 318 |
| Issue number | 3-4 |
| DOIs | |
| State | Published - Feb 15 2003 |
| Externally published | Yes |
Bibliographical note
Funding Information:We would like to acknowledge S. Friend and colleagues at Rosetta Inpharmatics [1] for making their database publicly available for the scientific community. We also thank X. He for performing the χ 2 test of gene expression distributions. Research at Eötvös University was supported by the Hungarian National Research Grant Foundation (OTKA), and by the Department of Energy and the National Institute of Health at the University of Notre Dame and at Northwestern University.
Keywords
- Analysis of graph topology
- Bioinformatics
- MRNA expression