Figure 3From: Predicting transcriptional regulatory interactions with artificial neural networks applied to E. coli multidrug resistance efflux pumpsOutput histograms of the feed-forward (FF) and bi-fan (BF) neural network predictors. Histogram for each type of feature vector, and the output of the classification neuron falls in the range [-1, 1]. The closer to -1, the higher the chances to be distinguished as a motif pattern. ANN classifiers were able to better discriminate whether motif samples were represented using hybrid correlation measures, as shown in BFall, BFskp, FFall and FFskp configurations.Back to article page