Figure 3. Computational tools to display gene expression profile data of five-paired cases of oral cancer. (A) Scatter plot identifies outlying genes between normal and malignant oral keratinocytes. (B) Hierarchical cluster analysis done on intensity values standardized by dividing by root mean square. Cosine correlation of similarity coefficient and complete linkage clustering classified the samples as shown. Normal ‘A’; Tumor ‘B’. (C) Self-organizing map (SOM) group expressed genes into co-expressed clusters (GeneCluster). (D) Principal component analysis (PCA) identifies the most significant expression patterns in all the genes examined. (E, F) Screen shoot of GeneSpring displays of selected gene expression (E, ß-actin; F, GRO1), comparing normal and tumor oral keratinocyte expression. Note that expression of ß-actin is relatively similar between normal and tumor specimens as well as between samples. GRO1, on the other hand, is dramatically overexpressed in malignant oral keratinocytes.