Fig. 4: Predictive performance for misaligned train and test cohorts. | Communications Biology

Fig. 4: Predictive performance for misaligned train and test cohorts.

From: Improving genetic risk prediction across diverse population by disentangling ancestry representations

Fig. 4

Prediction accuracy (AUC) for different models (PRS-Clumping trained on European (PRS-Clumping EUR), PRS-Bayes trained on European (PRS-Bayes EUR), PRS-Lasso trained on European (PRS-Lasso EUR), PRS-Bayes trained on African (PRS-Bayes AFR), PRS-Lasso trained on African (PRS-Lasso AFR), Neural Network trained on European (NN-EUR), Neural Network trained on African (NN-AFR), and DisPred) on the practical setting of when there is alignment or mismatch between the ancestry of training and test participants. All prediction models are trained on ADSP (EUR participants on the left and AFR participants on the right).

Back to article page