用于寡核苷酸二级结构预测的热力学数据库研究进展

Research progress of the thermodynamic database for oligonucleotide secondary structure prediction

基于核酸分子杂交的生物技术(如PCR)在病原微生物检测、临床诊断等诸多领域中应用广泛,此类技术的可靠性在于寡核苷酸分子与其靶点结合的高稳定性与特异性,而精确预测寡核苷酸与靶分子结合的二级结构是分析其稳定性与特异性的关键。其中,基于热力学的最近邻模型是寡核苷酸二级结构预测最为可靠的计算方法,但其精确性强烈依赖于精确的热力学参数。由于寡核苷酸分子二级结构的复杂性,除了完美匹配外,还需要错配、内环、膨胀环、末端摇摆、CNG重复、GU摆动等特殊结构的热力学数据。本文综述了近年来用于寡核苷酸二级结构预测的有效热力学数据库及相关计算方法,并指出当前热力学数据库的局限及未来发展方向。

The nucleotide hybridization based molecular biological technologies like PCR have been widely used in many fields, such as pathogenic microorganism detection, clinical diagnosis. And the accurate prediction of secondary structures between oligonucleotide and its binding sites is the key to these technologies. The Nearest-Neighbor Model based on thermodynamics is the most accurate method to predict oligonucleotide secondary structure, and the precision mainly depends on the thermodynamic parameters. Meanwhile, the diversity of secondary structure requires different thermodynamic parameters for different motifs, including perfect matches, mismatches, internal loops, bulge loops, dangling ends, CNG repeats, and GU wobble base pairs. Therefore, this review summarized the current parameter sets available for oligonucleotide secondary structure prediction. We also pointed out the limitations and future development directions of the thermodynamic database.