DNA Microarray Technology and Data Analysis in Cancer Research

The following sections are included:

• Preface

• Introduction of Authors

• Contents

The following sections are included:

• Experimental Procedure

• Experimental Design

• Quality Control

• Interpretation of DNA Microarray Data

• Advantages and Disadvantages

The following sections are included:

• Solid Tumors

• Blood Cancers

• Our DNA Microarray Study Using Mouse Model of BCR-ABL-Induced Leukemia
  • Leukemia mouse model study
    • Rationale
    • Experimental design
    • Results
  • Cell line study

The following sections are included:

• Experimental Design

• Method
  • Robust multi-chip averaging (RMA)
  • iterPLIER

• Quality Control Diagnostics
  • Saturation
    • Raw intensities for arrays containing a saturated probe
  • Transformed intensities across arrays
  • Normalized intensities across arrays
  • Scatterplot of normalized intensities
  • Average MA plot of normalized intensities

• Statistical Analysis
  • Analysis of variance (ANOVA) model
  • Contrasts
    • Fs permutation p-value distribution
    • p- and q-value summary
    • Volcano plot

• Coefficient of Variation Analysis

The following sections are included:

• Research Subjects

• Experimental Design
  • Rationale

• Fold Change Analysis

• More Information on SDL Global Optimization
  • Genetic algorithms (GAs)
  • SDL global optimization algorithms

The following sections are included:

• Leukemia Cell Line Study
  • Introduction
  • Datasets
  • Analysis strategies
    • Data sorting based on 20 chromosome groups
    • Dynamics of gene expression
    • Absolute difference between P190 and P210
    • Prefiltering process (gene selection)
    • Grouping genes based on the deviations
    • Analyzing the selected 8832 genes
    • The second sharp jump of deviations
    • Top 13 genes which are most differentially expressed across P190 and P210
  • Discussion and conclusion

• Analyses of Publicly Available Human Microarray Data
  • Introduction
  • Datasets
    • Colon data
    • Leukemia data

• Overall Methodology
  • Orthogonal arrays (OAs) and sampling procedure
  • Objective function
  • Search space reduction for global search
  • Mathematical form of SDL optimization
    • Definition of local minima
    • Definition of global minima
    • How to find the global minima
  • Multi-subset class predictor
  • Validation (predicting through a voting mechanism)

• Experimental Results

• Discussion

• Conclusion

As emphasized in Chapter 1, DNA microarray technology allows the parallel and simultaneous detection of more than 30000 genes in cells. Although large genome-scale cDNA or RNAi screens are powerful and efficient, they examine only one gene at a time, and will not uncover biological activities that often rely on multiple collaborating genes. Thus, DNA microarray is one of the best assays available for studying complex biological processes at a transcriptional level. At present, DNA microarray technology in cancer research is at a stage that, at best, provides information about disease-associated molecular signatures derived from analysis of the expression of basically all genes. There is still a long way to go before a diagnostic decision can be made based solely on DNA microarray data. This is largely due to the lack of a more powerful method for analysis of DNA microarray data. The development of new analytical methods of DNA microarray data is a critical step to take. In this regard, our SDL global optimization method described in this book has significantly changed our way of analyzing DNA microarray data, and has brought us much closer to making decisions based on microarray data…

The following sections are included:

• References

• Index