Skip to main content


Figure 1 | BMC Research Notes

Figure 1

From: Fast accurate missing SNP genotype local imputation

Figure 1

Average imputation accuracies of the seven methods, fastPHASE (fPH), NN, WNN, SVM, MC, NeuralNet (NeuN) and BaseLine (BL), on the three different density level human datasets: density-0.01, density-0.1 and density-1. The general tendencies are, on the low density datasets, no methods performed significantly better than BaseLine, which is a simple majority vote; on the median density datasets, methods started to perform better than BaseLine, with fastPHASE being the best and our methods NN and WNN and SVM performing not significantly different; on the high density datasets, the differences between fastPHASE and the second group, between the second and the third group of MC and NeuralNet, and between the third group and BaseLine became significantly larger.

Back to article page