Drug design is a creative process that combines different scientific expertise. With the development of increasingly powerful computers, disciplines such as molecular modeling and, in particular, drug design, are becoming an important component of drug discovery. However, modern software often limits the user interaction with the computer calculation, reducing the potential for researchers to use their knowledge in the design process. For this reason, interactive methodologies have been investigated in recent years. In particular, haptic-driven simulators offer the possibility for users to drive and control the modeling simulations, efficiently combining human knowledge and computational power. In this article, we will discuss the state-of-the-art and future perspectives of such methodologies.

In this article, we describe an improved cell-list approach designed to match the Kepler architecture of General-purpose graphics processing units (GPGPU). We explain how our approach improves load balancing for the above algorithm and how warp intrinsics are used to implement Newton's third law for the nonbonded force calculations. We also talk through our approach to exclusions handling together with a method to calculate bonded forces and 1-4 electrostatic scaling using a single Cuda kernel. Performance benchmarks are included in the last sections to show the linear scaling of our implementation using a step minimization method. In addition, multiple performance benchmarks demonstrate the contribution of various optimizations we used for our implementations. © 2013 Wiley Periodicals, Inc.

In this paper, we describe a hybrid meta-heuristics method of energy minimization and conformational sampling and its application into our haptic-driven molecular modelling simulator. The proposed method has been designed to suit real-time molecular docking simulations, where the time-lapse between two successive ligand poses is relatively small. In these situations, the energy minimization problem becomes increasingly complex and chaotic. The algorithm is tuned to take advantage of recent advances in GPU computing with asynchronous kernel execution, which has allowed us to include full protein flexibility in the real-time interactive, haptic-driven simulations. Finally, in this paper, we will also discuss the implementation of such high-performance computing approaches in our software, discussing the results of our initial validation studies, highlighting the advantages and limitations of such interactive methodology.