Related Experiment Videos
DNA microarray normalization methods can remove bias from differential protein expression analysis of 2D difference
David P Kreil1, Natasha A Karp, Kathryn S Lilley
1Department of Genetics/Inference Group (Cavendish Laboratory), University of Cambridge, Cambridge, UK. D.Kreil@gen.cam.ac.uk
Bioinformatics (Oxford, England)
|March 27, 2004
Summary
This study validates differential protein expression analysis using two-dimensional Difference Gel Electrophoresis (DIGE). A novel normalization method using a variance stabilizing transform and Z-score improves accuracy for analyzing protein expression data.
Area of Science:
- Proteomics and Bioinformatics
- Quantitative protein expression analysis
Background:
- Two-dimensional Difference Gel Electrophoresis (DIGE) enables high-throughput protein expression analysis.
- Limited studies exist on the validity of differential protein expression analysis and normalization methods in DIGE.
- Understanding experimental variance effects is crucial for accurate protein quantification.
Purpose of the Study:
- To systematically assess the impact of random experimental variance on differential protein expression analysis using DIGE.
- To investigate and compare the effectiveness of different normalization methods for pooled DIGE data.
- To develop a robust method for determining reliable, gel-independent signal thresholds.
Main Methods:
- Performed multiple 'same-same' DIGE comparisons to evaluate experimental variance.
- Applied a variance stabilizing transform (adapted from DNA microarray analysis) combined with a robust Z-score.
- Compared the proposed normalization method against established proteomics normalization techniques.
Main Results:
- Significant fluctuations in spot intensities across gels necessitate robust normalization.
- The proposed method, accounting for dye-specific background and intensity scale differences, established reliable signal thresholds.
- Established proteomics normalization methods failed to yield cross-validation-consistent thresholds.
Conclusions:
- Accurate normalization is essential for reliable differential protein expression analysis in DIGE.
- The developed normalization strategy, integrating a variance stabilizing transform and Z-score, provides gel-independent thresholds.
- This approach enhances the validity and reproducibility of DIGE-based quantitative proteomics.