Related Experiment Videos
Clustering of gene expression data using a local shape-based similarity measure
Rajarajeswari Balasubramaniyan1, Eyke Hüllermeier, Nils Weskamp
1Max-Planck Institute for Terrestrial Microbiology, Department of Organismic Interactions Karl-von-Frisch-Strasse, 35043 Marburg, Germany.
Bioinformatics (Oxford, England)
|October 30, 2004
Summary
We developed CLARITY, a new method for analyzing gene expression data from microarray time course experiments. CLARITY effectively clusters genes with similar expression patterns, revealing co-regulated genes and biological functions.
Area of Science:
- Bioinformatics
- Computational Biology
- Genomics
Background:
- Microarray technology facilitates large-scale gene expression studies.
- Analyzing gene expression profiles helps identify co-regulated genes.
- Biological relationships often manifest as similar, time-shifted expression patterns.
Purpose of the Study:
- To introduce CLARITY (Clustering with Local shApe-based similaRITY), a novel method for analyzing time-course microarray data.
- To identify co-regulated genes and functional relationships using gene expression profiles.
Main Methods:
- CLARITY employs a local shape-based similarity measure using Spearman rank correlation.
- The method is inspired by the BLAST algorithm for sequence alignment.
- It does not require data normalization and is robust to noise.
Main Results:
- CLARITY successfully clustered gene expression time series from yeast (Saccharomyces cerevisiae) cell cycle data.
- Detected similar and time-shifted expression sub-profiles.
- Clustered genes showed significant enrichment with shared or related functions based on MIPS classification.
Conclusions:
- CLARITY is an effective tool for analyzing time-course microarray data.
- The method can identify biologically relevant gene clusters based on expression patterns.
- CLARITY aids in understanding gene co-regulation and functional relationships.