Skip to main content
. 2022 Mar 5;434(11):167530. doi: 10.1016/j.jmb.2022.167530

Figure 1.

Figure 1

An illustration of the PEPPI pipeline. This pipeline functions by analyzing a pair of input sequences via a series of independent modules, including structure similarity, sequence similarity, neural network classification, and functional association data. These modules are combined using a naïve Bayesian consensus classifier, which provides the final interaction score as a log-likelihood ratio.