PopART Simulation to Model the Transmission of HIV in Zambia and South Africa

人类免疫缺陷病毒（HIV）是一种攻击人体免疫系统的病毒。虽然目前没有有效的治愈方法，但有有效的治疗方法（例如抗逆转录病毒治疗）使感染者能够活得更长、更健康。目前的目标是让95%的艾滋病毒感染者了解自己的艾滋病毒状况，95%的确诊艾滋病毒感染者接受持续的抗逆转录病毒治疗，95%接受抗逆转录病毒治疗的人接受抗逆转录病毒治疗，病毒得到抑制。由于非洲是受影响最严重的地区，占全球艾滋病毒新发感染总数的三分之二，因此艾滋病毒研究的重点是降低非洲的流行率和发病率。我们希望扩展预先开发的PopART-IBM模型，这是一个计算效率很高的基于个体的模型，用于模拟艾滋病毒在非洲高流行社区的传播。在轮换项目中，我增加了女性性工作者结构和卡波特韦药物成分，以模拟基于当前世界的更真实的场景。然后我们可以比较结果，并为政府或当地社区提供减少艾滋病毒传播的建议。目前的模型有几个限制，例如，固定的风险组（低、中、高）决定了一个人可以拥有的最大数量的伴侣。因此，需要更多的灵活性和变化，以便从分布中得出这个数字。在未来，我们将首先推导出数学公式，并将其实现为进一步校准的代码。然后我们可以用新的参数进行模拟，这将更好地模拟现实社会。该项目属于EPSRC人工智能和工程、卫生和政府数据科学（ASG），因为我们拥有的原始数据是调查数据。它相当混乱，需要一些数据科学技术来确定哪些字段对提取更重要。这个项目属于EPSRC生物信息学，因为研究目标是非洲的艾滋病毒传播，使用的方法是数学和计算建模。该项目属于EPSRC临床技术，在我们的模型中，正常背景艾滋病毒检测，社区艾滋病毒护理提供者提供的额外家庭干预试验和卡波特韦药物都是模拟的。我们还将进行敏感性分析，为诊所提供建议，说明覆盖这些检测的适当比例，以及哪些人属于需要随访的关键人群。该项目属于EPSRC全球挑战研究基金，EPSRC确保更好的健康，老龄化和福祉，因为我们的目标研究领域是非洲的艾滋病毒，其中大多数是发展中国家。我们的目标是让更多的人意识到他们的艾滋病毒状况和种子治疗，从而抑制病毒，过上更健康的生活。这个项目属于EPSRC数学生物学、统计学和应用概率，因为我们的模型涉及数学推导，例如，使用幂律分布来模拟合作伙伴的数量。这个项目属于EPSRC软件工程，因为我们通过计算模拟来构建模型，为了更好的效率和维护，它是用C语言编写的。

Human immunodeficiency virus (HIV) is a virus that attacks the body's immune system. Although there's no current effective cure, there are effective treatments (e.g. antiretroviral therapy) to enable the infected people to live a longer and healthier life. The current aim is to let 95% of all people living with HIV know their HIV status, 95% of all people with diagnosed HIV infection receive sustained ART and 95% of all people receiving antiretroviral therapy receiving ART have viral suppression. As Africa is the most affected region, accounting for two-thirds of the global total of new HIV infections, the HIV research focus is to reduce the prevalence and incidence in Africa.We want to extend the pre-developed PopART-IBM model, which is a computationally efficient individual-based model to simulate HIV transmission in African communities with high prevalence. In the rotation project, I added the female sexual worker structure and cabotegravir drug component to simulate more realistic scenarios based on the current world. Then we can compare the results and also give suggestions to the government or local community to reduce HIV transmission. There are several limitations in the current model, for instance, the fixed risk groups (low, medium and high), which determine the maximum number of partners an individual could have. Consequently, more flexibility and variation are desired, so that this number should be drawn from a distribution. In the future, we will derive the mathematical formulas first and implement them into codes for further calibration. Then we can run simulations with the new parameters, which would better mimic real society.This project falls within the EPSRC AI and Data Science for Engineering, Health and Government (ASG) as the raw data we have is survey data. It is quite messy and requires some data science techniques to determine which fields are more important for extraction. This project falls within the EPSRC Biological Informatics as the research target is HIV transmission in Africa and the used approach is mathematical and computational modelling. This project falls within the EPSRC Clinical technologies as in our model, normal background HIV testing, extra household-based intervention trials offered by community HIV care providers and cabotegravir drug are all simulated. We will also carry out a sensitivity analysis to provide clinics advice about what's the proper proportion to cover these tests and who is in the key population to follow up.This project falls within the EPSRC Global Challenges Research Fund, EPSRC securing better health, aging and wellbeing as our targeted research area is HIV in Africa, where most are developing countries. We aimed at making more people aware of their HIV status and seed treatment to become virally suppressed for healthier lives.This project falls within the EPSRC Mathematical Biology, Statistics and applied probability as our model involves mathematical derivation, for instance, using power law distribution to simulate the number of partners.This project falls within the EPSRC Software engineering as we build the model by computational simulation, which is written in C for better efficiency and maintenance.