项目名称: 面向不同对称性分子的自适应高性能单颗粒重构算法研究
项目编号: No.61502475
项目类型: 青年科学基金项目
立项/批准年度: 2016
项目学科: 计算机科学学科
项目作者: 王功明
作者单位: 中国科学院生物物理研究所
项目金额: 20万元
中文摘要: 对称自适应函数(Symmetry Adapted Function, SAF)算法是效果最好的单颗粒重构算法,生物学家采用该算法获取多种接近原子分辨率的分子结构,辅助生物领域的前沿科学研究。但是,该算法在某些方面还存在不足,而且仅在20面体分子重构中取得成功,尚未报道其它对称性分子重构的真实结果。本项目解决SAF算法在20面体分子重构中存在的问题,并将优化后的SAF算法拓展到非20面体分子。主要内容有:(1)提出适用于20面体分子的高性能SAF算法,解决缔合勒让德多项式计算、球面SAF矩阵插值等方面的问题。(2)设计适用于非20面体分子三维重构的傅里叶-贝赛尔算法,作为非20面体分子SAF算法的性能参照对象。(3)将适用于20面体分子的高性能SAF算法拓展到非20面体分子,开发出适用于不同对称性分子的自适应高性能单颗粒重构软件包,为研究分子结构和功能的关系提供更精确的信息。
中文关键词: 冷冻电镜;分子结构;高性能;自适应;对称性
英文摘要: In theory, SAF(Symmetry Adapted Function) algorithm is the best in all single particle reconstruction algorithms. Biologists use this algorithm to obtain a variety of molecular structures at near-atomic resolution, which facilitates the advanced science research in the field of biology. However, SAF algorithm has self-shortcomings in some aspects, only is succeed in the Cryo-EM data of icosahedral molecule, and the reconstruction case of non-icosahedral molecule has not reported. In our study, the problems of SAF algorithm for icosahedral molecule are solved, and the optimized SAF algorithm is extended to the non-icosahedral molecule. The details are as follows. First of all, the high-performance SAF algorithm for icosahedral molecule is proposed, and the problems such as associated Legendre polynomial calculation and spherical SAF matrix interpolation are solved. After that, the Fourier-Bessel algorithm for non-icosahedral molecule is proposed and taken as the reference object of SAF algorithm. Subsequently, the high-performance SAF algorithm for icosahedral molecule is extended to non-icosahedral molecule and compared with the Fourier-Bessel algorithm. Finally, an self-adaptive and high-performance single particle reconstruction software package is developed, which is used to reconstruct the different symmetrical molecules automatically. In sum, our method can supply more accurate information for discovering the relationship between molecular structure and function.
英文关键词: Cryoelectron Microscopy;Molecular Structure;High Performance;Self-Adaptive;Symmetry