Cryptosporidium Artificial Intelligence Network Analysis of Drug Action (CANADA)

将一种新药推向市场可能需要长达12年的时间，耗资26亿美元(过去10年增长了140%)，药物开发总成本高达600亿美元/年。这远远不是可持续的或可获得的，并造成了一种经济障碍，阻止制药公司投资于非有利可图但非常重要的研究领域，包括大流行预防和抗菌素耐药性领域，目前和未来都会给社会带来重大成本。[《用于公益药物发现的人工智能》，GPAI，2021]。在这个项目中，我们分享了消除这一障碍的长期愿景，分享了麦肯锡的愿景，即迫切需要“全面转变新药开发的研发--更快地500天，更好地适应患者需求，并降低25%的成本”。考虑到这一长期目标，该项目专注于开发一种概念验证，以克服药物开发过程中的瓶颈，即测试用于寄生虫病的新化合物。传统上，这一测试过程是劳动和资源密集型的。提出的解决方案是开发一个基于人工智能的药物设计系统，该系统自动预测现有化合物对蛋白质-蛋白质相互作用网络的影响。该系统将使用机器学习算法来分析蛋白质之间的相互作用，并预测药物作用。通过了解这些相互作用，该系统将能够识别能够有效地针对网络中的关键基因的化合物，同时将毒性降至最低。该项目将专门专注于在隐孢子虫寄生虫上测试这一基于人工智能的药物设计系统，隐孢子虫寄生虫是一种胃肠道寄生虫，会导致腹泻、营养不良，有时还会导致死亡，特别是在儿童中。这个项目使用的假设和方法是基于首席调查员(PI)以前进行的研究，并将使用经过验证的生物学方法进一步完善和测试。这个项目的最终目标是开发一个系统，可以预测哪些化合物在治疗寄生虫病方面最有效，对宿主的毒性最低

Bringing a new drug to market can take up to twelve years and cost 2.6 billion USD (a 140 percent increase in the past ten years), with a total drug development cost up to $60 billion/year. This is far from being sustainable or accessible and creates an economic barrier that prevents pharmaceutical companies from investing in non-lucrative and yet very important "fields of research in domains including pandemic prevention and antimicrobial resistance, with major current and future costs for society". ["Artificial Intelligence for Public Good Drug Discovery", GPAI, 2021]. In this project, we share the long-term vision of removing this barrier, sharing the McKinsey vision of an urgent need to "transform R&D for new-drug development holistically-500 days faster, better tailored to patient needs, and 25 percent cheaper".With this long-term goal in mind, this project focuses on developing a proof-of-concept for overcoming the bottleneck in the drug development process, which is the testing of new compounds for parasitic diseases. Traditionally, this testing process has been labour and resource intensive. The proposed solution is to develop an AI-based drug design system that automates the process of predicting the effect of existing compounds on protein-protein interaction networks. This system will use machine learning algorithms to analyse the interactions between proteins and predict the drug action. By understanding these interactions, the system will be able to identify compounds that can effectively target key genes in the network while minimising toxicity. The project will specifically focus on testing this AI-based drug design system on the Cryptosporidium parvum parasite, which is a gastrointestinal parasite which causes diarrhoea, malnutrition, and sometimes death particularly in children. The hypotheses and methods used in this project are based on previous studies conducted by the Principal Investigators (PIs) and will be further refined and tested using proven biological approaches.The ultimate goal of this project is to develop a system that can predict which compounds will be the most effective in treating parasitic diseases with the minimum levels of toxicity to the host