The Computational Pharmacology Core B (PI: Ivet Bahar, J.K. Vries Chair of Computational and Systems Biology, and Associate Director of U of Pittsburgh Drug Discovery Institute) will provide expertise in developing and applying computational tools for modeling protein-inhibitor interactions, assessing target proteins, pathways and networks, quantitative evaluation of target systems pharmacological features, and analyses of high throughput and high content screening (HTS/HCS) data. The Core resources for high-throughput virtual screening of molecular libraries as well as modeling protein-ligand interactions and mining databases will be extensively used to serve the research goals of the Center. The specific aims will focus on two areas: (i) quantitative and systems pharmacology (OSP) analyses and (ii) ligand- and target-based discovery approaches. QSP will deliver new therapeutical targets with small-molecule modulators and polypharmacology strategies for experimental tests in Project 1. We will identify new potential targets by analyzing RNAi and small-molecule screening data from Project 2 using machine learning techniques. We will also assess the polypharmacological effect of simultaneously targeting more than one protein. In particular, we will investigate the effects of combining modulators of autophagy and proteasome pathways, newly identified targets, and/or other specific targets investigated in Project 1 towards designing polypharmacological strategies. Target- and ligand based approaches, on the other hand, will be used for repurposing existing drugs, computational assessment of identified targets, and identifying new compounds. We will mine drug-target association databases such as DrugBank, and protein structural databases using sequence- and binding site shape similarity metrics. Targets with known structural data will be examined for availability of alternative more selective inhibition mechanisms by identifying potential allosteric sites. To this aim, we will use the structural ensemble analysis software and elastic network models developed in the Bahar lab. Finally, we will identify new chemical scaffolds by pharmacophore-based screening of purchasable compound libraries. Structure-based models of Project 1 targets, experimental HTS hits from Project 2, compounds from literature (e.g. autophagy inducers), and chemical mitogens investigated in Project 3 will be used in pharmacophore model building. Compounds that we identify and prioritize will be tested in cell lines, mouse and worm models of ATD. Results will be used to refine pharmacophore models and to iteratively search for new compounds

It is now widely established that progress in drug discovery requires thorough computer-aided analyses of high throughput screening data from experiments, in addition to computer-aided studies for identifying pathways and target proteins, and prioritizing small molecules. The Computational Pharmacology Core will assist the Center Projects 1-3 by providing expertise in these areas.