To derive a mutational set that would be simultaneously predictive of MSI and low aneuploidy, different sets predicting MSI and low aneuploidy were obtained independently. The genetic algorithm described above was applied to the cluster of gastrointestinal and endometrial tumor samples using (a) 100 repetitions aiming to maximize the Spearman correlation coefficient ρ between each set and the aneuploidy level and (b) 100 repetitions aiming to maximize the performance (AUC of ROC curve) of each set in predicting the MSI status. Mutations significantly selected for both tasks (with combined X2 P-value < 0.1 for the selection scores over 100 repetitions for both (a) and (b)) were chosen to compose the final MSI-aneuploidy set.
Do you have any questions about this protocol?
Post your question to gather feedback from the community. We will also invite the authors of this article to respond.