首页|香荚兰挥发性成分色谱保留的理论预测模型

香荚兰挥发性成分色谱保留的理论预测模型

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
香荚兰是一种具有实用价值的一种名贵天然香料，还具有极高的药用价值。为建构香荚兰提取物挥发性成分保留指数神经网络的定量结构-保留相关性模型，根据香荚兰提取物挥发性成分中原子的特性和连接状态，在分子拓扑理论基础上，提出了一种新的分子结构参数-分子键指数mF，并计算了香荚兰提取物挥发性成分的连接性指数(mX)，优化筛选了其中的3X和4Xpc，将选取的分子键指数和连接性指数的4种分子结构参数，作为人工神经网络方法的输入层变量，色谱保留指数作为输出层变量，采用4-10-1的神经网络结构，构建的预测模型总的相关系数R为0。996 3，根据该模型计算得到的香荚兰提取物挥发性成分的保留指数预测值与实验值的平均相对误差仅为1。26％，该误差明显低于多元回归方法预测值的平均相对误差4。71％。结果表明，香荚兰提取物挥发性成分的保留指数与4种分子结构参数之间具有很好的非线性关系，杂原子及—CH3，—CH2—，—CH(\)和(C)等基团片段，是影响香荚兰提取物挥发性成分保留指数大小的主要因素。
Theoretical prediction model of chromatographic retention for volatile components in Vanilla
Vanilla is a valuable natural fragrance with practical importance and possesses significant medici-nal value.To construct a neural network model of the structure-retention relationship for the chromato-graphic retention index of volatile components of vanilla extract,a novel molecular structure parameter-molecular bond index mF was derived based on characteristics and connectivity of the atoms in volatile components using molecular topology theory.The molecular connectivity index(mX)of volatile compo-nents in vanilla was calculated,3X and 4Xpc were optimized screened.The four types of molecular struc-ture parameters of the molecular bond indices and connectivity indices were used as input layer variables of the artificial neural network method,while the chromatographic retention index was used as the output layer variable.Besides,the 4-10-1 neural network structure was adopted and the artificial neural network method was used to establish a neural network model.The total correlation coefficient(R)is 0.996 3.The average relative error between the predicted value and the experimental value of the retention index was only 1.26％,which was substantially lower than the average relative error of 4.71％using multiple regression methods.The results showed that there is a strong non-linear relationship between the chroma-tographic retention index of volatile components from Vanilla and the four molecular structure parameters./Heteroatom and functional groups such as—CH3,—CH2—,—CH(\),(C)are the main factors influ-encing the chromatographic retention index of volatile components from Vanilla extract.

vanillavolatile componentschromatographic retention indexmolecular bond indexquantita-tive structure-retention relationshipneural network

堵锡华、李靖、陈艳、宋明

展开 >

徐州工程学院材料与化学工程学院,江苏徐州 221018

香荚兰 挥发性成分 色谱保留指数 分子键指数 定量结构-保留相关 神经网络

Jiangsu Natural Science FoundationJiangsu Provincial Natural Science Foundation Major ProjectScience and Technology project of Xuzhou

BK2017116818KJA430015KC21286

2024

10.13563/j.cnki.jmolsci.2023.09.004

分子科学学报
中国化学会

分子科学学报

CSTPCD
影响因子：0.434
ISSN：1000-9035
年,卷(期)：2024.40(1)
  • 25