Data-driven engineering of novel drugs by machine-learning-derived understanding of the molecular basis for drug action

该项目将为基于人工智能的数据驱动的药物“工程”铺平道路，这将塑造新一代更有效的药物，副作用更少。G蛋白偶联受体（GPCR）是约三分之一药物的靶点，但这些受体通常会激活许多独立的理想和不理想的信号通路，降低治疗效果并增加不良副作用。为了设计更好的药物，我们必须更详细地了解药物作用的机制。在该项目中，我们将开发最先进的人工智能和机器学习方法，将来自高通量丙氨酸扫描的独特途径激活数据与来自分子化合物库高通量筛选的结构信息和数据相结合，以了解药物作用的分子基础GPCR中，预测GPCR类别中大量药物的作用，并预测将产生所需信号传导谱的配体。

This project will pave the way for Artificial Intelligence-based data-driven drug "engineering" that will shape a new generation of more efficacious drugs with fewer side-effects. G protein coupled receptors (GPCRs) are the target for around one-third of drugs, but these typically activate a number of separate desirable and undesirable signalling pathways, reducing the efficacy of the therapy and increasing undesirable side effects. To design better drugs, we must understand the mechanisms of drug action in much greater detail. In this project, we will develop state-of-the art AI and machine learning approaches to combine unique pathway activation data from high-throughput alanine scanning with structural information and data from high-throughput screens of molecular compound libraries to to understand the molecular basis of drug action in GPCRs, predict the action of a large range of drugs across GPCR classes, and predict ligands that will give rise to a desired signalling profile.