Fig_S2_NN_Model

An artificial NN model showing input neurons characterizing each AHE locus, the single hidden neuron layer (H1), and the output variable predicting RF distances between each gene tree and the species tree (RFgtst). This diagram shows both intercept terms (bias), B1 and B2, which increases NN efficiency, and synaptic weights indicated by the thickness and bolding of lines connecting the neurons