To increase the speed and efficiency of knowledge dissemination, the computational and informatics community has proposed the use of nano-publications - modular, nimble units of communication designed to be machine-readable. Nano-publications as currently defined by the World Wide Web Consortium (W3C), are validated, core scientific statements, with associated context. However, implementation and adoption of the nano-publication has thus far faced many challenges. Common strategies to generate nano-publications have focused on extracting them from the existing literature in a “top-down” fashion, by means of either manual curation or automated text mining. Neither method is sustainable, thus limiting the use of this top down approach. Using existing tools for text mining the large collection of data in a traditional journal article results in a data analysis burden that is unmanageable. Moreover, existing methods directed at the traditional article can result in predictions that are difficult to properly curate and validate.