Streamlined and Targeted Design of NOTUM Inhibitors as Colorectal Cancer Drugs

肠癌是全球主要的死亡原因，仅在英国每年就有1.7万人死亡。现有的治疗方法受到耐药性发展的影响。Kuano正在寻求将量子模拟和人工智能的创新结合起来，以生产针对一种最近发现在癌症发展中发挥关键作用的酶(Notus)的新一代药物。像许多疾病一样，肠癌的一个主要原因是功能失调或酶活性过高--这些化学机器加速了所有生物的化学作用。这就是为什么30%的现有药物针对酶。Kuano平台自动化了一种被称为过渡态药物(TSD)设计的成熟方法，该方法特定于酶。为了加速化学反应，酶以一种称为过渡态的构象将其输入的化学物质极其紧密地结合在一起--反应的速度取决于结合的强度。大多数靶向酶的药物通过结合到与输入化学物质相同的位置来阻止它们的功能-如果我们使用过渡态作为模板，我们可以设计出结合更强的药物。每种酶都进化了一个独特的过渡状态，这一事实也意味着这个模板允许我们使我们的药物更具选择性(因此毒性更低)，并且对酶的微小变化很健壮，否则可能会导致耐药性。药物开发是一项耗时、昂贵的工作，通常需要15年时间和大约20亿英镑才能将一种新的治疗方法投入临床。人工智能方法有助于降低成本，但基于使用之前测试的分子的信息，限制了创新。传统的TSD设计涉及昂贵的靶向特定实验，这些实验需要多年才能针对特定的酶进行量身定制。通过在计算机上进行实验，我们可以极大地减少过程中涉及的成本和时间线，同时还可以利用与标准人工智能方法相关的所有节省。更好的是，通过使用最先进的模拟来了解药物靶标的化学行为，我们不仅限于以前确定的起点，还可以探索所有的化学空间来寻找最好的药物。Kuano展示了他们的平台可以成功地复制现有的TSD，并在实验室中验证了它可以开发用于药物设计的新型起始化合物。这个项目将使我们能够生产出新的抗癌药物候选药物，这将代表着医学科学的巨大突破，并最终为社会带来重大突破。

Bowel cancer is a leading cause of death worldwide, with 17,000 deaths each year in the UK alone. Existing therapies suffer from the development of resistance. Kuano is looking to apply an innovative combination of quantum simulation and AI to produce a new generation of drugs targeting an enzyme (NOTUM) recently discovered to play a key role in cancer development. Like many diseases, a major cause of bowel cancer is malfunctioning or overactive enzymes - the chemical machines that accelerate chemistry within all living things. This is why 30% of existing drugs target enzymes. The Kuano platform automates a well established approach known as transition state drug (TSD) design which is specific to enzymes. To accelerate chemical reactions, enzymes bind their input chemicals incredibly tightly in a conformation known as the transition state - the speed of the reaction depending on the strength of the binding. Most drugs targeting enzymes block their function by binding to the same site as the input chemicals - if we use the transition state as a template we can design drugs that bind much more strongly. The fact that each enzyme has evolved a unique transition state also means that this template allows us to make our drugs more selective (and hence less toxic) and robust to small changes in the enzyme that might otherwise lead to resistance. Drug development is a time consuming, expensive endeavour, typically taking 15 years and around £2 billion to get one new treatment into the clinic. AI approaches are helping to reduce costs but are based on using information from previously tested molecules, limiting innovation. Traditional TSD design involves expensive and target specific experiments which take many years to tailor to a specific enzyme. By conducting experiments in a computer we can vastly reduce the cost and timelines involved in the process while also taking advantage of all of the savings associated with standard AI approaches. Even better, by using state-of-the-art simulations to understand the chemical behaviour of drug targets we are not limited to previously identified starting points and can explore all of chemical space to find the best drugs. Kuano demonstrated their platform can successfully reproduce existing TSDs and verified in the laboratory it can develop novel starting compounds for drug design. This project will allow us to produce new cancer drug candidates that would represent a huge breakthrough in medical science and, ultimately, to society.