Quality Assessment of Microarray Data
Author Information
Author(s): John McClure, Ernst Wit
Primary Institution: University of Glasgow
Hypothesis
Post-normalization checks of microarray data are necessary to identify potential problems.
Conclusion
The study presents methods for assessing the quality of microarray data after normalization to identify issues that could affect analysis results.
Supporting Evidence
- Post-normalization checks can identify clerical mistakes, hybridization problems, normalization issues, and mishandling problems.
- Dimension reduction techniques and visualization methods can help in assessing the quality of microarray data.
- Problems identified can either be rectified or excluded from the data analysis.
Takeaway
This study shows that checking microarray data after it's been processed is important to find mistakes that could lead to wrong conclusions.
Methodology
The study uses dimension reduction techniques, false array plots, and correlograms to identify problems in microarray data.
Potential Biases
The exploratory nature of the methods means they do not formally test for a problem-free state of the data.
Limitations
The methods discussed are not exhaustive and may not cover all potential issues with microarray data.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website