论文部分内容阅读
解链温度预测在引物和探针设计中具有重要的作用,本研究以384条寡核苷酸的解链温度数据为材料,随机分为训练集(279条)和测试集(69条)样本,利用训练集样本对建立的GRNN人工神经网络进行训练;再利用训练好的人工神经网络对测试集样本的解链温度进行预测,发现本研究所建立的GRNN人工神经网络的平均预测误差为2.44±0.98℃,最大误差为5.77℃,说明本研究建立的GRNN人工神经网络具有较好的预测性能,完全可以用于寡核苷酸解链温度的预测。同时比较了GRNN人工神经网络与目前常用的3种邻近法在预测寡核苷酸解链温度上的差异,发现Breslauer(1986)建立的预测方法误差较大,其平均误差为6.81±3.90℃,Santalucia(1996)建立的预测方法次之,平均误差为2.41±1.96℃,Sugimoto(1996)建立的预测方法最准确,其平均误差为1.57±0.96℃,分析了各种预测方法产生误差的原因,为今后开发新的寡核苷酸解链温度预测工具提供了新的思路和方法。
Prediction of melting temperature plays an important role in primer and probe design. In this study, 384 oligonucleotide melting temperature data were randomly divided into a training set (279) and a test set (69) , Training samples were used to train the established GRNN artificial neural network, and then the trained artificial neural network was used to predict the melting temperature of the test sample. The average prediction error of the GRNN artificial neural network established in this study was 2.44 ± 0.98 ℃, the maximum error is 5.77 ℃, indicating that the GRNN artificial neural network established in this study has good predictive performance and can be used for the prediction of oligonucleotide melting temperature. At the same time, we compared the difference of the predicted melting temperature of the three oligonucleotides between the GRNN artificial neural network and the three commonly used methods, and found that the prediction error established by Breslauer (1986) was larger with an average error of 6.81 ± 3.90 ℃, The prediction method established by Santalucia (1996) is second, with an average error of 2.41 ± 1.96 ° C. The prediction method established by Sugimoto (1996) is the most accurate with an average error of 1.57 ± 0.96 ° C. The causes of the errors in various prediction methods are analyzed, It provides a new idea and method for the future development of a new oligonucleotide melting temperature prediction tool.