CAREER: Deep representation learning for exploration and inference in biomedical data

生物系统本质上是复杂的。越来越复杂的技术被用于生物医学科学，以理解这种复杂性，并了解导致疾病的潜在因素。这些技术以多种不同的形式产生大量数据，从基因和蛋白质在单个细胞中表达方式随时间的变化，到大型患者群体的详细临床成像数据和数十万人的全基因组测序研究。这些新开发的数据库可以帮助揭示支撑健康和疾病的重要机制和途径。然而，这些数据集中包含的信息与提取有意义见解的能力之间存在很大差距。在这里，PI建议通过开发基于数学基础的新机器学习方法来解决这个问题，这将使我们能够理解这些复杂的数据集。PI将开发深度表征学习技术，专注于全面了解数据的结构，动力学，相互作用和预测特征，并将允许有关在不同背景下驱动疾病的潜在调控机制的特定假设。该提案还将涉及培训一名博士后、研究生和指导当地高中生。此外，该项目还将推动在线研讨会的发展，以便向来自全国各地的各种参与者广泛传播无监督数据分析的知识。该项目提出通过三个主要方面推进生物医学数据分析。第一个重点是基于数据几何、图形信号处理和拓扑概念，结合强大的深度学习系统，形成数据的深度多尺度表示。这种表示将允许以可扩展的方式探索数据的结构和有意义的预测性抽象。我们的第二个重点是整合多种数据模式，并使用最佳传输和生成模型组织大量相关数据集，以深入了解整个患者队列或扰动条件。我们的第三个重点是使用神经网络（随机微分方程）和图形ODE（常微分方程）网络来学习数据的高维随机动力学，以深入了解潜在的基因调控网络。我们在几个特定的生物医学挑战的背景下应用我们的方法。实现这些目标将有助于整合和探索大量数据，以解释潜在的监管机制和动态表型变化。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Biological systems are inherently complex. Increasingly sophisticated technologies arebeing used in biomedical science in order to make sense of this complexity and to understand theunderlying factors that cause disease. These technologies generate vast amounts of data in manydifferent forms, from changes in how genes and proteins are expressed in individual cells over time,to detailed clinical imaging data on large patient populations and whole genome sequencing studiesacross hundreds of thousands of people. These newly developed datatypes could help uncoverimportant mechanisms and pathways that underpin health and disease. However, there is a largegap between the information contained in these datasets and the ability to extract meaningfulinsights. Here the PI proposes to address this by developing new machine learning approaches basedon mathematical foundations that will allow us to make sense of these complex datasets. The PI will develop deep representation learning techniques that focus on gaining overall insight into thestructures, dynamics, interactions, and predictive features of the data, and will allow specific hypotheses regarding the underlying regulatory mechanisms that drive disease in differentcontexts to derived. The proposal will also involve training a postdoc, graduate student, and mentorship of local high school students. In addition, it will enable the development of an online workshop towidely disseminate knowledge of unsupervised data analysis to a diverse array of participants fromacross the country.This project proposes to advance biomedical data analysis via three main thrusts. The first thrust is focused on forming deep multiscale representations of the data based on data geometry, graph signal processing, and topological concepts, in combination with powerful, deep learning systems. Such representations will allow for exploration of structure and meaningful, predictive abstractions of the data in a scalable fashion. Our second thrust is focused on integrating multiple modalities of data and organizing multitudes of related datasets using optimal transport and generative models to gain insight into entire cohorts of patients or perturbation conditions. Our third thrust is focused on learning high dimensional stochastic dynamics of the data using neural SDE (stochastic differential equation) and graph ODE (ordinary differential equation) networks to gain insight into underlying gene regulatory networks. We apply our approaches in the context of several specific biomedical challenges. Achieving these aims will enable integration and exploration of a large volume of data for explaining underlying regulatory mechanisms and dynamic phenotypic changes.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.