Skip to main content
Fig. 3 | BioData Mining

Fig. 3

From: Genetic risk score for ovarian cancer based on chromosomal-scale length variation

Fig. 3

The receiver operating characteristic curves for different model predictions. The area under the curve for the Gradient Boosting Machine model was 0.89. An actual predictive test for ovarian cancer would require choosing a threshold. Depending on the threshold, the true positive rate and false positive rate (or equivalently the sensitivity and specificity) will vary. This graph demonstrates how these two factors will vary

Back to article page