Table 2 Classification results observed by cross validation using SVM classifiers. Figures represent achieved classification accuracies, i.e. the fraction of samples correctly classified. The upper table shows results for cross validation analysis of both data sets of a pair, where samples for training and testing are selected randomly from both studies. For this, data sets were integrated by either MRS or QD. The bottom table contains the results of a cross-validated classification analysis performed separately on each study, using all available gene expression data after pre-processing (without applying MRS or QD). Abbreviations: MRS, median rank scores; QD, quantile discretization, SVM, support vector machine.
both data sets integrated | Â | Â |
|---|---|---|
| Â | MRS | QD |
Prostate cancer | 97.67 % | 97.56 % |
Breast cancer | 87.01 % | 88.97 % |
Acute myeloid leukemia | 90.60 % | 90.20% |
original data | Â | Â |
Prostate cancer | Dhanasekaran et al. | Welsh et al. |
| Â | 95.28 % | 99.09 % |
Breast cancer | Gruvberger et al. | West et al. |
| Â | 80.52 % | 86.73 % |
Acute myeloid leukemia | Bullinger et al. | Valk et al. |
| Â | 68.53 % | 99.90 % |