Related Experiment Videos
Spurious spatial periodicity of co-expression in microarray data due to printing design
Gábor Balázsi1, Krin A Kay, Albert-László Barabási
1Department of Pathology, Feinberg School of Medicine, Northwestern University, Ward Building 6-204, 303 East Chicago Avenue, Chicago, IL 60611, USA. g-balazsi@northwestern.edu <g-balazsi@northwestern.edu>
Nucleic Acids Research
|July 31, 2003
Summary
Analyzing transcriptome data can be skewed by experimental biases. This study reveals how microarray probe arrangement affects mRNA expression data and introduces a method to filter these biases for more accurate biological insights.
Area of Science:
- Molecular Biology
- Bioinformatics
- Systems Biology
Background:
- Global transcriptome data analysis is crucial for understanding cellular function.
- The impact of experimental biases on transcriptome data accuracy is not well understood.
- Microarray technology is widely used for measuring mRNA expression levels.
Purpose of the Study:
- To investigate the influence of microarray spatial arrangement and printing on mRNA expression data.
- To develop a method for filtering technology-derived biases from transcriptome data.
- To improve functional predictions derived from transcriptome datasets.
Main Methods:
- Analysis of log-ratio data from Saccharomyces cerevisiae cell cycle experiments.
- Numerical method development to identify and remove spatial and printing biases.
- Comparison of functional predictions before and after bias correction.
Main Results:
- Spatial probe arrangement and printing procedures significantly impact mRNA expression log-ratio data.
- A novel numerical method successfully filters out technology-derived contributions.
- Bias correction leads to enhanced accuracy in functional predictions.
Conclusions:
- Experimental biases inherent in microarray technology can obscure true biological signals.
- A systematic approach to identify and compensate for such biases is essential.
- Improved data analysis methods enhance the reliability of transcriptome-based functional predictions.