Forecasting trajectories of HIV transmission networks with a novel phylodynamic and deep learning framework

总结 尽管出现了联合抗逆转录病毒疗法，但目前的艾滋病毒流行病仍然无法预防， 旨在显著降低全球流行率和发病率的干预战略。为了 实现2020年联合国艾滋病规划署90-90-90的目标（90%的艾滋病毒感染者被诊断，90%的人被诊断 接受持续抗逆转录病毒治疗，90%的人接受治疗以维持病毒抑制）， 必须开发创新工具，用于预测本地化的增长和轨迹， 由特定传播集群驱动的亚流行病。系统动力学分析已广泛用于 艾滋病毒领域，以追踪起源和重建病毒的人口历史，在地方，区域和全球 水平然而，这类研究到目前为止只是回顾性的，几乎没有或根本没有能力做出预测 未来的流行趋势。该项目的总体目标是开发一种创新的 将动态推理和行为网络数据与人工 智能算法能够预测艾滋病毒传播集群的未来轨迹，并提供关键信息， 新感染的决定因素。我们建议通过实现三个具体目标来实现这一目标：1.发展 一个基于生物动力学的PRIDE模块，用于预测艾滋病毒感染热点[感染者]; 2.发展一种行为 基于网络的艾滋病毒感染风险PRIDE模块[未感染者]，以及3.开展焦点小组活动， 将新的PRIDE预测技术部署到公共卫生领域，并通过 对等变革代理模型。特别是通过与佛罗里达卫生部的密切合作， （FLDoH），我们将分析FLDoH在过去12年中收集的现有数据库 包括广泛的HIV分子序列、临床和行为网络数据。佛罗里达有一个艾滋病病例 2016年，每10万人中有24.0人死亡，目前是美国每年死亡人数第三多的州。 发病率。我们与FLDoH的合作伙伴关系将确保拟议研究的结果将用于 通过优化基于公共卫生的监测方案， 采取更有效的预防措施。

SUMMARY Despite the advent of combined antiretroviral therapy, the ongoing HIV epidemic still defies prevention and intervention strategies designed to reduce significantly both prevalence and incidence worldwide. In order to achieve the 2020 UNAIDS 90-90-90 goal (90% of people living with HIV diagnosed, 90% of people diagnosed to be on sustained antiretroviral treatment, and 90% of people on treatment to maintain viral suppression), it is necessary to develop innovative tools that can be used for predicting the growth and trajectory of localized sub-epidemics driven by specific transmission clusters. Phylodynamic analysis has extensively been used in the HIV field to track the origin and reconstruct the virus demographic history both at local, regional and global level. However, such studies have been so far only retrospective, with little or no power to make predictions about future epidemic trends. The overarching goal of the prosed project is to develop an innovative computational framework coupling phylodynamic inference and behavioral network data with artificial intelligence algorithms capable of predicting HIV transmission clusters future trajectory, and informing on key determinants of new infections. We propose to achieve this goal by carrying out three specific aims: 1. Develop a phylodynamic-based PRIDE module to forecast HIV infection hotspots [the infected]; 2. Develop a behavioral network-based PRIDE module for risk of HIV infection [the uninfected], and 3. Carry out focus groups for deploying the new PRIDE forecasting technology into public health, and implement prevention through the peer change agent model. In particular, through a close partnership with the Florida Department of Health (FLDoH), we will analyze existing databases that the FLDoH has assembled over the past twelve years including extensive HIV molecular sequence, clinical and behavioral network data. Florida had an HIV case rate of 24.0 per 100,000 people in 2016, and it is currently the third state in the USA in terms of yearly incidence. Our partnership with the FLDoH will ensure that the results of the proposed research will be used to curtail the HIV epidemic by optimizing public health based surveillance programs, informing targeted intervention strategies, and implementing more effective prevention measures.