Skip to main content

Table 1 Detailed configurations of the method

From: Deep mixed model for marginal epistasis detection and population stratification correction in genome-wide association studies

Collector (1D-CNN)Convolutional layerNum. of Kernels: 16Kernel Size: 1000 x 1Padding: Same
  Initializer: Truncated normal initializerActivation: ReLU
 Pooling layerSize: 2000Stride: 2000 
 1st fully-connected layerOutput: 32Dropout rate: 0.9 
 2nd Fully-connected layerOutput: 1  
Selector (LSTM)Weighting layerNum. of units: p (one-to-one layer)
 Hidden layerNum. of units: 0.15p
Optimizer (ADAM)Learning rate: 0.001Batch size: 128  
Other hyperparamsCollector’s epoch: 20Selector’s epoch: 1500  
  1. The architecture and hyperparameters are selected through the experiments with simulated data, and are used without changes for real data experiments