Improving the Prediction of Complex Diseases by Testing for Multiple Disease-Susceptibility Genes

摘要: Studies have argued that genetic testing will provide limited information for predicting the probability of common diseases, because incomplete penetrance genotypes and low magnitude associated risks general population. Such studies, however, usually examined effect one gene at time. We argue disease prediction multifactorial diseases is greatly improved by considering multiple predisposing environmental factors concurrently, provided model correctly reflects underlying etiology. show how likelihood ratios can be used to combine from several tests compute developing a disease. To concurrent use improves disease, we logistic regression with simulated case-control data hypothetical influenced risk factors. As practical example, also apply this approach venous thrombosis, nongenetic Under reasonable conditions, markedly For panel three (factor V Leiden, prothrombin variant G20210A, protein C deficiency) increases positive predictive value thrombosis least eightfold. Multiplex has potential improve clinical validity diseases.