偏最小二乘回归分析法在黄芩抑菌谱-效关系及数学模型研究中的应用

以黄芩药材指纹图谱及其对金黄色葡萄球菌抑菌率为基础数据，采用偏最小二乘回归分析法（PLS）进行数据的谱-效相关性分析，并建立数学模型。结果表明，黄芩中有5个成分对抑菌率的影响较大，建立模型的药效预测值偏差在10%以内，能够较好地预测药效。该方法及所建立的数学模型能成为预测黄芩抑制金黄色葡萄球菌的药效作用的一种有效的手段。

Application of partial least squares regression analysis in the study of the antibacterial spectrum-effect relationship and mathematical model of Scutellariae Radix

Based on the fingerprint of Scutellariae Radix and its inhibition rate of Staphylococcus aureus, the partial least squares regression method was used to analyze the correlation of the antibacterial spectrum-effect relationship, and a mathematical model was established. Results showed that five components in Scutellariae Radix have a significant effect on the bacteriostasis rate. The established model could predict the efficacy of Scutellariae Radix with a deviation of less than 10%, indicating the desired prediction for the efficacy. The partial least squares regression method and the established mathematical model can provide an effective way to predict the inhibition rate of Staphylococcus aureus for the efficacy of Scutellariae Radix.