A plethora of disease-related information can be found in the biomedical literature. However, many of these sources are unstructured and are not designed with searching in mind. SciBite has developed a method that combines Semantic Analytics and Machine Learning to unlock the potential of biomedical literature and quantify disease similarities based on their phenotypic signatures. Our method involves mining the scientific literature for potential phenotype similarities, ranking those similarities, and creating an integrated knowledge-base for visual and computational exploration. Combining our technique with SciBite's ChemAxon integration further improves the potential of this knowledge-base for performing drug discovery and repurposing at-scale.