The target prediction score is given by our target prediction model which is constructed based on naïve Bayes method. The target prediction score is just combined likelihood ratio (LR) of multiple pieces of supporting evidence, according to the naïve Bayes rule. Generally combined LR>1 indicates that there is evidence supporting the protein to be a target. The larger the combined LR of the protein is, the more properly the protein can be used as a drug target for a certain type of drug. See Documents of POPPIT for more details.