Skip to main content


Fig. 1 | Genome Medicine

Fig. 1

From: Predicting cancer type from tumour DNA signatures

Fig. 1

Performance of different classifiers. Using (a) only somatic point-mutated genes, (b) only copy number altered genes and (c) both somatic point-mutated genes and copy number altered genes as the predictors. The mean overall accuracy, with its 95 % confidence interval band, was computed using the results from 50 sets of randomly subsampled training data and their corresponding test data. For SVM-RFE and random forest, we first ranked the genes in decreasing order of their importance, before using an increasing number of them to train and test the classifiers. For L 1-logistic regression, we varied the parameter λ to control the number of genes selected. The accuracy of a random classifier is also plotted to provide a baseline for comparison. The random classifier assigns a tumour sample to the different cancer classes with probabilities proportional to the size of those classes in the training data set

Back to article page