Fig. 2From: MUMAL2: Improving sensitivity in shotgun proteomics using cost sensitive artificial neural networks and a threshold selector algorithmArchitecture of the artificial neural network used in MUMAL and in this work. It illustrates the input layer (six nodes), the hidden layer (four nodes by default), and the output layer (one node), where the activation function is the sigmoid to map the ANN’s output into a value in the range [0, 1]Back to article page