Targeting AKAP12/Gravin Using Cloud-Computing for Drug Discovery and Molecular Simulation

Emerging technology’s newest evolution, “Cloud-Computing”, provides opportunities to study pharmacological leads and mechanisms. Cloud-Computing can be viewed as an abstraction to traditional computer clusters. It implements a conceptual layer that divides the hardware from the software. Cloud providers like Microsoft-Azure, Amazon-AWS, and Google-Cloud-Platform allow Virtual Machines running Linux to be configured as abstract clusters tailored for our applications and sharing resilient data with a RESTful (Representational-State-Transfer) architecture that integrates our high performance computational pipeline with web services. This project optimizes our drug discovery pipeline, utilizing several programs; including AutoDock, Chimera, Firefly and GROMACS, in the “Cloud” The goal of this project is to understand protein functional mechanisms at a molecular level and to exploit them for chemical lead development using docking simulations with the millions of ligands from computational libraries. One of the big data frameworks being used for this new pipeline is Apache Spark’s PySpark implementation, which distributes data over the cluster with DataFrames (highly parallel data ), allowing workload processing many times faster than traditional methods. In order to maximize utilization of these tools, we are using Microsoft Azure services to set up and run powerful Virtual Machines capable of executing many cores and offering even more scalability. These powerful tools offer the ability to run computational-intensive programs, such as GROMACS or AutoDock, over many cores simultaneously and process data to iterate on lead optimization. AKAP-12/Gravin is critical in cardiac function through B2-adrenergic receptor, Protein-Kinase-A, and Protein-Kinase-C signaling and here we using docking to explore receptor inhibition.