In silico safety pharmacology

点击翻译按钮获取中文摘要

PROJECT SUMMARY: A major factor plaguing drug development is that there is no drug-screening tool that can distinguish between drugs that will induce cardiac arrhythmias from chemically similar safe drugs. The current approaches rely on substitute markers such as action potential duration or QT interval prolongation on the ECG. There is an urgent need to identify a new approach that can predict actual proarrhythmia from the drug chemistry rather than relying on surrogate indicators. We have brought together an expert team to innovate at the interfaces of experimental and computational modeling disciplines and develop an in silico simulation pipeline to predict cardiotoxicity over multiple temporal and spatial scales from the atom to the cardiac rhythm. An essential and unique aspect of our approach is that we propose to utilize atomistic scale simulation to predict the transition rates of ion channels and adrenergic receptors and how they are modified by drug interaction. We hypothesize that it is the subtleties of these interactions that are likely to be the critical determinants of drug associated safety or proarrhythmia. In the last award period, we successfully developed an unprecedented linkage: We connected the highly disparate space and time scales of ion channel structure and function. We utilized atomistic simulation to compute drug kinetic rates were directly used as parameters in a hERG function model. The model components were then integrated into predictive models at the cell and tissue scales to expose fundamental arrhythmia vulnerability mechanisms and complex interactions underlying emergent behaviors. Human clinical data were used for model validation and showed excellent agreement, demonstrating feasibility of this new approach for cardiotoxicity prediction. In this renewal application we propose to hugely extend this approach to include prediction of the interaction of cardiac channel gating and drug interaction as well as the inclusion of adrenergic receptor interactions with drugs. Another essential aspect of safety pharmacology is the development of new approaches to allow more efficient drug design, screening and prediction of cardiotoxicity. Therefore, we will seek to develop, extend and apply a variety of machine learning and deep learning approaches to improve drug discovery by predicting proarrhythmia from the drug chemistry with an efficient process that identify drug congeners via machine learning to maximize therapy and minimize side effects. Finally, we propose to classify drugs into categories based on proarrhythmia risk in normal and diseased virtual tissue settings. The multiscale model for prediction of cardiopharmacology that we will develop in this application will be applied to projects demonstrating its usefulness for efficacy or toxicity of drug treatments in the complex physiological system of the heart.