Torsion angle dynamics for NMR structure calculation with the new program DYANA
Torsion angle dynamics for NMR structure calculation with the new program DYANA
复制标题
DOI：
10.1006/jmbi.1997.1284
复制
发表时间：
1997-10-17
影响因子：
5.6
通讯作者：
Wuthrich, K
中科院分区：
文献类型：
--
作者：
Guntert, P;Mumenthaler, C;Wuthrich, K
由核磁共振实验收集的距离约束和扭角约束计算蛋白质和核酸三维结构的新程序Dyana(动力学算法)在扭角空间进行分子动力学模拟退火法，并使用快速递归算法对运动方程进行积分。在笛卡尔坐标空间中，扭角动力学可以比分子动力学更有效，因为减少了自由度，并且没有伴随的高频键和角振动，这允许在结构计算中使用更长的时间步长和/或更高的温度。这也代表了在扭转角度空间的可变目标函数法与前身程序DIANA所使用的REDAC策略相比的重大进步。在用于表示核磁共振结构的“束”中，每个被接受的构象的Dyana计算时间与其他目前可用的结构计算算法相比是有利的，当使用DEC Alpha 8400 5/300计算机时，对于165个氨基酸残基的蛋白质，计算时间约为160秒。从具有随机扭转角值的构象开始的测试计算进一步表明，Dyana能够有效地计算高质量的蛋白质结构和核酸结构，这些结构的氨基酸残基高达400个。(C)1997年学术出版社有限公司。
The new program DYANA (DYnamics Algorithm for Nmr Applications) for efficient calculation of three-dimensional protein and nucleic acid structures from distance constraints and torsion angle constraints collected by nuclear magnetic resonance (NMR) experiments performs simulated annealing by molecular dynamics in torsion angle space and uses a fast recursive algorithm to integrate the equations of motions. Torsion angle dynamics can be more efficient than molecular dynamics in Cartesian coordinate space because of the reduced number of degrees of freedom and the concomitant absence of high-frequency bond and angle vibrations, which allows for the use of longer time-steps and/or higher temperatures in the structure calculation. It also represents a significant advance over the variable target function method in torsion angle space with the REDAC strategy used by the predecessor program DIANA. DYANA computation times per accepted conformer in the ''bundle'' used to represent the NMR structure compare favorably with those of other presently available structure calculation algorithms, and are of the order of 160 seconds for a protein of 165 amino acid residues when using a DEC Alpha 8400 5/300 computer. Test calculations starting from conformers with random torsion angle values further showed that DYANA is capable of efficient calculation of high-quality protein structures with up to 400 amino acid residues, and of nucleic acid structures. (C) 1997 Academic Press Limited.