The Healthspan Machine: an automated method to screen for interventions that slow ageing

人们的寿命变长了，但许多人几年来健康状况不佳，这给NHS、家庭和整个社会带来了越来越大的负担。对衰老生物学的更好理解将使研究人员能够设计干预措施，使老年人保持更长时间的流动性和总体上的健康。一个人保持健康的时间被称为他们的“健康持续时间”，我们的目标是找到延长这一健康时间的干预措施。老龄化领域中一些最有成效的研究使用了小型实验室动物，因为它们衰老得很快，我们可以弄清楚它们的基因和环境条件是如何影响衰老的。线虫秀丽线虫的寿命只有几周，对这种动物的研究已经在衰老生物学以及许多其他生物学领域产生了几个见解。使用这种蠕虫进行生物研究的优势之一是，它可以用于基因筛查。基因筛查涉及通过搜索大量的突变蠕虫来寻找那些不同的蠕虫。研究这些特定的突变体揭示了使它们不同的基因的功能。然而，找到一个长寿的突变体是非常困难的，因为即使在基因相同的动物之间，衰老也是非常不同的，而且需要对蠕虫进行几周的观察。这项建议解决了这些问题。首先，我们不会寻找存活时间更长的蠕虫，而是寻找那些活动时间更长的蠕虫，即那些具有更长“健康寿命”的蠕虫。其次，我们将使用自动化技术来测量蠕虫的移动。我们的建议是受哈佛医学院最近开发的“寿命机器”的启发。他们的方法使用特别改装的高规格平板扫描仪来测量大量蠕虫的寿命。虽然这项技术可行，并已被许多研究小组采用，但它对用户来说可能是具有挑战性的，而且成本高昂，而且它消耗相当大的空间和能源。我们的建议通过使用几个小型零售摄像头来跟踪蠕虫的运动来克服这些问题。“寿命机”可以检测到每条蠕虫的死亡。我们的方法是监测一群蠕虫随着年龄的增长是如何减速的。我们已经修改了运行摄像头的软件，以便在短时间内拍摄一组蠕虫的数百张图像。使用最初开发的监测天文图像的技术，我们处理蠕虫图像，并使用处理后的图像来测量蠕虫的运动。定期监测蠕虫群将使我们能够测量随年龄增长的运动量下降。通过扩大这项技术的规模，我们可以测试数千种不同的基因和条件，这些基因和条件可能会延缓这种运动的衰退。我们已经制作了这个“Healthspan机器”的原型，用两个摄像头对准两个含有蠕虫的培养皿。要扩展到使用100多个摄像头的功能正常的机器，存在许多障碍，但我们设计了一套解决方案来克服这些挑战，并制定了机器开发的合理顺序。例如，我们需要能够非常快地存储和处理大量数据，我们计划通过使用与摄像头通信的计算机网络来实现这一点。我们将首先将6到12个摄像头连接到计算机，然后将多台计算机连接在一起。每台计算机将处理来自所连接的摄像头的原始图像，生成更小的文件，以便进一步的数据处理和存储更容易。到项目结束时，我们的目标是拥有一台功能正常的机器，可以用于许多不同的筛选实验，并被世界各地的其他研究人员使用。这些实验将帮助我们了解导致长寿的基因和环境条件。然后，我们还可以使用“Healthspan Machine”来筛选化合物，寻找新的营养和药物干预措施，使我们更长时间地保持健康。

People are living longer but many suffer several years of ill health leading to an increasing burden to the NHS, families and society in general. A greater understanding of the biology of ageing would allow researchers to design interventions that would keep the elderly mobile and generally healthy for longer. The time that someone stays healthy is termed their "healthspan", and we aim to find interventions that extend this time of health. Some of the most productive research in the field of ageing has used small lab animals because they age quickly and we can work out how their genes and environmental conditions influence ageing. The nematode worm Caenorhabditis elegans has a lifespan of only a few weeks, and studies with this animal have produced several insights in the biology of ageing, as well as in many other areas of biology. One of the strengths of using this worm for biological research is that it can be used for genetic screens. A genetic screen involves searching through a large number of mutant worms to find those that are different. Studying these particular mutants reveals the function of genes that make them different. However, finding a long-lived mutant is very difficult because even between genetically identical animals, ageing is very variable and the worms need to be observed over several weeks. This proposal addresses these problems.Firstly, rather than looking for worms that live longer, we will find those that stay moving for longer, i.e. those with a longer "healthspan". Secondly, we will use automated techniques to measure worm movement. Our proposal is inspired by the "Lifespan Machine", which was recently developed at Harvard Medical School. Their method uses specially adapted high-specification flatbed scanners to measure the lifespan of large numbers of worms. While this technology works and has been taken up by a number of research groups, it can be challenging and expensive for users to implement and it consumes considerable space and energy. Our proposal overcomes these issues by using several small retail cameras to track worm movement. The "Lifespan Machine" detects the death of each worm. Our approach is to monitor how a group of worms slow down as they age. We have modified the software that runs the cameras so that hundreds of images of a group of worms are taken in a short space of time. Using techniques originally developed to monitor astronomical images, we process the worm images and use the processed images to measure the movement of the worms. Monitoring the groups of worms at regular time intervals will allow us to measure the decline in movement with age. By scaling up this technology we can test 1000s of different genes and conditions that might delay this decay of movement. We have generated a prototype of this "Healthspan Machine" with two cameras trained on two petri dishes containing worms. There are many obstacles to scaling up to a functioning machine using 100+ cameras but we have devised a set of solutions to overcome these challenges and a logical order of developing the machine. For example, we need to be able to store and process a large amount of data very quickly and we plan to do this by using a network of computers that communicate with the cameras. We will start with linking up 6 to 12 cameras to computers, and then link multiple computers together. Each computer will process the raw images from the attached cameras, producing smaller files so that further data processing and storage is easier. By the end of the project, we aim to have a functioning machine that can be used for many different screening experiments and used by other researchers across the world. These experiments will help us understand the genes and environmental conditions that lead to a long healthspan. We can then also use the "Healthspan Machine" to screen compounds for new nutritional and pharmaceutical interventions that keep us healthier for longer.