Application of the classifier to CNVs obtained in routine diagnostics of patients with mental retardation.

The accuracy of the classifier developed was tested on an independent cohort of CNVs. Phase 1 contained the validation set of 1,203 CNVs known to be either rare de novo or commonly inherited. 43 of the 49 rare de novo CNVs known to be associated with MR were correctly classified, and 1,085 of the 1,154 common inherited CNVs known to be benign were correctly classified, thus giving an overall classification accuracy of 94%. The false positive rate was 0.05 and the false negative rate was 0.12. The positive predictive value was 0.38 and the negative predictive value was 0.99. Phase 2 consisted of the application set containing 94 CNVs of unknown clinical significance. Of the 41 rare inherited CNVs the classifier identified 27 CNVs as MR-associated and 14 as being benign. 53 candidate CNVs for which the inheritance could not be determined were also classified, from which 46 were classified as being MR and 7 CNVs were classified as being benign.