Over the past decade, the drug discovery landscape has changed dramatically. Many pharmaceutical companies have shifted their research efforts from enzyme targets to programs based on more complex phenotypic screens. These screens create rich datasets, which often lack the resolution found in traditional enzyme assays. Even in cases where drug targets are known, teams must design compounds that are selective for the target of interest and avoid potential off-target liabilities. In addition, drug discovery programs often track dozens of assays, creating large volumes of data that must be analyzed and understood. Public repositories containing millions of chemical structures associated with tens of millions of biological activity values are now readily available. While this data could provide a benefit to ongoing drug discovery programs, its use is currently limited. This changing landscape has created a number of new challenges and opportunities for those engaged in Computer-Aided Drug Design. These challenges require us to move beyond typical application areas and develop new tools and techniques.This conference will focus on many of the new challenges facing drug design, and seek to set new research directions to address these challenges. We will address a number of pertinent topics including: The application of computational methods to the design of biologics Data analysis and visualization techniques to support multi-objective optimization The application of computational methods to drug discovery programs driven on data from phenotypic assays The impact of open source software and open data on drug design The impact of "big data" on drug discovery Computational methods applied to the optimization of binding kinetics In addition, we will examine case studies from existing drug discovery programs to identify new areas where computation can make an impact.