In synthetic biology, designing a new genetic construct demands in-detail studies of its candidate components individually and in a composition with each other. These costly wet lab experiments require considerable amount of time and usually result in undesired output. In this paper, we propose a method for the extraction of existing or novel synthetic devices from the available biological parts or devices from iGEMs BioParts Registry and ordered the resulting devices based on their computed reliabilities. This method is very efficient and it helps the wetlab biologists in designing their genetic devices based on the given input and output in a reasonable amount of time. This method is implemented in "Mining for BioBricks" (M4B), a web-based application that facilitates the prediction of novel genetically made devices based on the given input and output. © 2012 IEEE.