Dimensionality Reduction and Search for Analyzing Protein Structure

描述（由申请人提供）：几乎不可能找到在某个阶段不主动涉及单个蛋白质或蛋白质复合物的生理功能。我们对蛋白质结构如何与蛋白质功能或大分子复合物中的合作机制相关的知识存在巨大差距，尽管这些问题多年来一直在挑战科学家。了解蛋白质和蛋白质复合物等系统如何工作也有很大的希望，这将影响药物和酶的合理设计，并为某些疾病提供更好的治疗方法。 该项目的智力价值在于开发一种新的框架来探索蛋白质和蛋白质复合物的构象空间。我们的目标是生成上述系统的几何上不同的低能构象。我们将追求降维方法的发展，这些方法适合于复杂的分子系统，并可用于表示这种系统的复杂性。使用这些表示，我们将积极探索蛋白质的构象景观，最初，蛋白质复合物，在稍后的阶段。我们相信，降维和有效的搜索算法的紧密耦合将导致一种方法，可以对大型系统的原因与一些概率的保证，目前难以捉摸的目标。我们的方法的一个显着特点是，我们将修改和更新我们的低维表示的探索的构象景观的进展，以最好地代表所考虑的系统，因为它的演变。我们工作的结果可用于研究生物大分子的可能形状，并阐明其功能。至于应用方面，我们将首先解决与分子对接和计算机辅助药物设计有关的问题。从长远来看，我们的目标是与实验学家合作研究分子机器和大分子组装。该项目的更广泛影响是通过以下方式实现的：（a）与德克萨斯医学中心的跨学科合作，这将影响应用和计算数学、计算机科学、生物学和生物化学的学生，（B）对本科生、研究生和博士后学生进行培训、指导和参与研究活动，（c）在莱斯大学开设课程，（d）指导女本科生学习计算机科学;（e）参加国家科学基金会资助的莱斯大学和休斯顿独立学区的一个方案，其目标是吸引高中女生进入她们人数不足的领域。

DESCRIPTION (provided by applicant): It is virtually impossible to find a physiological function that does not actively involve single proteins or protein complexes at a certain stage. There is a huge gap in our knowledge of how protein structure relates to protein function or to cooperative mechanisms in macromolecular complexes, although these issues have been challenging scientists for many years. There is also great promise that understanding how systems such as proteins and protein complexes work, will impact the rational design of drugs and enzymes, and suggest better treatments for certain diseases. The intellectual merit of this project lies in the development of a novel framework to explore the conformational space of proteins and protein complexes. Our goal is to generate geometrically-distinct low energy conformations of the above systems. We will pursue the development of dimensionality reduction methods that are tailored to complex molecular systems and can be used to represent such systems compactly. Using those representations we will aggressively explore the conformational landscape of proteins, initially, and protein complexes, at a later stage. We believe that the tight coupling of dimensionality reduction and efficient search algorithms will result in a method that can reason about large systems with some probabilistic guarantees, a presently elusive goal. A distinguishing feature of our method is that we will modify and update our low-dimensional representations as the exploration of the conformational landscape progresses in order to best represent the considered system as it evolves. The output of our work can be used to study the possible shapes of a biomacromolecule and shed light on its function. As far as applications are concerned, we will first tackle problems that relate to molecular docking and computer-assisted drug design. In the long run, our goal is to study molecular machines and macromolecular assemblies in collaboration with experimentalists. The broader impact of the project is implemented through (a) interdisciplinary collaborations with the Texas Medical Center which will affect students in applied and computational mathematics, computer science, biology, and biochemistry, (b) training, mentoring and involvement in research activities of undergraduate, graduate and postdoctoral students, (c) course development at Rice University, (d) mentoring of women undergraduate students in computer science, and (e) participation in an NSF funded program of Rice University and the Houston Independent School District whose goal is to attract high school girls to fields where they are underrepresented.