Analysis of DIGE data using a linear mixed model allowing for protein-specific dye effects
Morten Krogh1, Yingchun Liu, Sofia Waldemarson
1Computational Biology and Biological Physics, Department of Theoretical Physics, Lund University, Lund, Sweden. mkrogh@thep.lu.se
Proteomics
|November 6, 2007
Summary
Protein-specific dye effects in differential in-gel electrophoresis (DIGE) experiments are significant. A new linear mixed model, DIGEanalyzer, accounts for these dye biases for more accurate protein analysis.
Area of Science:
- Proteomics
- Biochemistry
- Bioinformatics
Background:
- Differential in-gel electrophoresis (DIGE) enables simultaneous analysis of three protein samples per gel using Cy2, Cy3, and Cy5 fluorescent dyes.
- Accurate quantification in DIGE experiments is crucial for reliable proteomic studies.
Purpose of the Study:
- To investigate and quantify protein-specific dye effects in DIGE experiments.
- To develop and validate a statistical model that accounts for dye biases in DIGE data analysis.
Main Methods:
- Development of a linear mixed-effects model to analyze DIGE data, incorporating protein-specific dye effects.
- Implementation of the model in a freely available Java software called DIGEanalyzer.
- Validation using three large-scale DIGE experiments (173, 64, and 24 gels) with DeCyder software for spot analysis.
Main Results:
- Significant dye effects were observed in 19-34% of proteins across the experiments.
- Dye effects exceeding a 1.4-fold change were found in 1-6% of proteins.
- Median dye effects ranged from 1.07-fold (Cy5 vs. Cy3) to 1.16-fold (Cy3 vs. Cy2), with a maximum of seven-fold.
Conclusions:
- Protein-specific dye biases are a notable factor in DIGE experiments and require statistical correction.
- The DIGEanalyzer software provides a robust method for accounting for these dye effects, improving the accuracy of DIGE-based proteomic studies.
