Inverse Problem for Estimating Structure of Biological Macromolecules from SAXS

利用 SAXS 估计生物大分子结构的反问题

基本信息

批准号：
8518388
负责人：
Jay Breidt
金额：
$ 25.12万
依托单位：
COLORADO STATE UNIVERSITY
依托单位国家：
美国
项目类别：
财政年份：
2010
资助国家：
美国
起止时间：
2010-08-20 至 2014-07-31
项目状态：
已结题

项目摘要

DESCRIPTION (provided by applicant): Technologies to rapidly obtain information about the structure of macromolecular complexes are needed to understand how these complexes function and, in particular, how aberrant interactions between molecules may result in human disease. Small-angle X-ray scattering is a relatively simple experimental method to obtain low-resolution information that offers advantages over methods such as X-ray crystallography and NMR. However, recovering information about the structure of a molecule from a scattering pattern is an extremely ill-posed and ill-conditioned mathematical inverse problem. In addition, evaluating mathematical models for a scattering pattern requires a computationally demanding, high dimensional integration. As a consequence of these difficulties, the current state of practice in SAXS reconstruction of molecular structure falls short of its potential as a scientific tool. The proposed research project is a multidisciplinary investigation into the inverse problem of determining structural information about complex bio-molecules from SAXS scattering patterns. In a close, well-established partnership with biochemists using SAXS in their research, a team of computational scientists, mathematicians, and statisticians will undertake a systematic investigation of the mathematical properties of the SAXS inverse problem using new statistical and mathematical tools, devise novel computational methods for computing solutions of the inverse problem for specific data sets, undertake an analysis quantifying the effects of model uncertainty, experimental error, and computational error on identifications made using SAXS data, and apply these methods to biomedically relevant experimental systems. The project is driven by specific research problems in biochemistry and structural biology, where extracting all available information from experimental data is essential to revealing low-resolution information for large macromolecules or complexes in solution. Such information advances understanding of the function of proteins and macromolecular assemblies in health and disease.

描述（由申请人提供）：需要快速获取有关大分子复合物结构的信息，以了解这些复合物如何发挥作用，尤其是分子之间的异常相互作用如何导致人类疾病。小角度X射线散射是一种相对简单的实验方法，可获得低分辨率信息，比X射线晶体学和NMR等方法具有优势。但是，从散射模式中恢复有关分子结构的信息是一个极度不良且条件不足的数学反问题。另外，评估散射模式的数学模型需要计算苛刻的高维积分。由于这些困难，分子结构的重建中的当前实践状态均未达到其作为科学工具的潜力。拟议的研究项目是对确定有关萨克斯散射模式复杂生物分子的结构信息的反问题的多学科研究。在与生物化学家的紧密合作伙伴关系中，使用SAXS在他们的研究中，计算科学家，数学家和统计学家团队将对SAXS逆问题的数学属性进行系统调查，使用新的统计和数学工具，并设计新颖的计算工具用于计算特定数据集的反问题解决方案的方法，进行分析，量化模型不确定性，实验误差和计算误差对使用SAXS数据的识别的影响，并将这些方法应用于生物医学相关的实验系统。该项目是由生物化学和结构生物学方面的特定研究问题驱动的，其中从实验数据中提取所有可用信息对于揭示溶液中大型大分子或复合物的低分辨率信息至关重要。这些信息进一步了解了蛋白质和大分子组件在健康和疾病中的功能的理解。