Skip to main content

Table 6 Table summarizes the accuracy of ANN, GRU, BILSTM, LSTM, and 1DCNN model for 3 SNPs

From: Eye-color and Type-2 diabetes phenotype prediction from genotype data using deep learning methods

Model, SNPs = 3 Activation Dropout Optimizer Batchsize Epochs Validation Accuracy
ANN Sigmoid 0.2 Adam 1 10 0.2 0.88
Sigmoid 0.2 Adam 10 20 0.4 0.89
Relu 0.3 RMSprop 15 50 0.3 0.89
Relu 0.3 SGD 1 50 0.2 0.89
GRU Sigmoid 0.2 Adam 1 10 0.2 0.895
Sigmoid 0.2 RMSprop 10 50 0.3 0.895
BILTM Sigmoid 0.2 Adam 1 10 0.2 0.895
Sigmoid 0.3 RMSprop 10 100 0.3 0.895
LSTM Sigmoid 0.2 Adam 1 10 0.2 0.9
1DCNN Sigmoid 0.2 Adam 1 20 0.2 0.88
Sigmoid 0.2 Adam 1 50 0.2 0.89
Softmax 0.3 RMSprop 20 100 0.3 0.89
Softmax 0.2 Adam 20 50 0.2 0.89
Relu 0.3 RMSprop 15 50 0.4 0.89
  1. LSTM performs well with an accuracy of 0.9%