( Version 1.10) in a one-class response f...)Differential Gene Expression Analysis of Pure Cell Populations from Frozen UterineTissue Using Laser Capture Microdissection (LCM) and cDNA Microarrays

The quantitated intensity data were transformed to M & A values by the following equation:

R is background-subtracted Cy5 (IM sample) intensity and G is background subtracted Cy3 (INTER sample) intensity. Transformed data were normalized using lowess function implemented in S-plus statistical software package. (User defined parameter f for LOWESS function was set at 0.2.) The normalized data were used for SAM analysis and clustering analysis.

3. Figure 2 shows the identification of the genes with significant changes in expression. The graph shows the scatter plot of the observed relative difference d(i) versus the expected relative difference d_E(i). The red spots (right upper quadrant) indicate the genes with higher expression at IM, while the green spots (left lower quadrant) indicate the genes with higher expression at INTER sites. The microarray data were normalized with lowess and visualized by MA scatter plot (Figure. 3). The horizontal axis shows the A value representing the signal intensity and the vertical axis shows the M value representing the gene expression difference. Representatives from significant genes determined by SAM are indicated with black arrows. Some non-significant genes with high M values are marked with red arrows.

4. Figure 4 depicts two-dimensional cluster analysis of the data. Using the information from Gene Ontology Consortium (www.geneontology.org), we associated the genes selected by SAM with biological functions.

Results and Discussion
To identify the molecular basis for blastocyst implantation in the uterus, we compared RNA of luminal epithelium from IM and INTER sites. Pure luminal epithelium was is
'"/>

Source: