Fig. 4From: A generative model for constructing nucleic acid sequences binding to a proteinThe procedure for computing the binding affinity of a sequence s to a target protein p. After computing DeepBind scores of 200,000 random sequences by a DeepBind model m for p, an empirical cumulative distribution function was derived from the DeepBind scores. The function is discrete, but seems continuous due to a large number of data points. The binding affinity of s to p is the probability that the DeepBind score of s would be higher than that of a random sequenceBack to article page