Mathematical Prediction of Nucleic Acids 3-D Structures Using Inter-Spin Distances and Nonlinear Least Squares Analysis

Jordan Journal of Biological Sciences

Samer I. Awad¹ 

关键词: DNA;    Nucleic Acid;    Biological Macromolecule;    Site-directed spin labels;    Electron paramagnetic resonance;    Least-squares;    Levenberg-Marquardt.;   

DOI  :  

学科分类：生物科学（综合）

来源: Hashemite University, Deanship of Research and Higher Studies

【 摘 要 】

Nucleic acids consist of several double helical arms (three to eight of them). These arms are connected at a junction point, with or without several unpaired bases in one or more of the different strands. Current structural information on several nucleic acids is still limited. Electron paramagnetic resonance (EPR) can provide distance measurements of site-directed spin labels (SDSL) applied to the double helical arms. These distances can be used to generate several quadratic equations that characterize the three-dimensional (3-D) structure of the nucleic acid. In this work, a nonlinear least squares algorithm is used to solve these equations along with molecular constraints simultaneously. The solution that this algorithm calculates can be used to create the predicted 3-D structure. The algorithm was tested using twenty-five cases for known DNA structures. Root-mean-square deviation (RMSD) was used in this study to evaluate the accuracy of the predicted structures. The calculated RMSD values had an average of 2.56 and a standard deviation of 1.65.

【 授权许可】

CC BY