A novel approach is introduced for the prediction of the psychotomimetic activity of the substituted phenethylamines as psychedelic drugs. In contrast to the available quantitative structure‐activity relationship (QSAR) methods where they need complex computer codes, unusual descriptors, and expert users, it requires the molecular structure of the desired phenethylamine derivative. The largest reported values for 116 different substituted phenethylamines are used to derive and test the new correlation by 93 and 23 experimental data, respectively. The new correlation is based on the contributions of specific isomers under certain conditions and halogen atoms. Cross validations of the new correlation corresponding to the coefficients of determination for leave‐one‐out (Q²_LOO) and the leave‐many‐out (Q²_LMO) are the same and equal to 0.895, which is close to the coefficient of determination (R²=0.900) of the training set. Further statistical parameters also confirm the goodness‐of‐fit, goodness‐of‐prediction, accuracy, and precision of the novel model. For eleven compounds where their outputs of external validation data set of psychotomimetic activity from multiple linear regression analysis (MLR) and non‐linear multiple regression analysis (MNLR) QSAR models were available, the values of root mean squared error (RMSE) of the new model, MLR and MNLR are 0.172, 0.624 and 0.539, respectively.

中文翻译：

---

一种评估取代苯乙胺类药物拟精神活性的简单方法

引入了一种新方法来预测取代的苯乙胺作为迷幻药物的拟精神活性。与需要复杂的计算机代码，不寻常的描述符和专业用户的可用定量结构-活性关系（QSAR）方法相比，它需要所需的苯乙胺衍生物的分子结构。116种不同取代的苯乙胺的最大报道值分别用于通过93和23个实验数据得出和测试新的相关性。新的相关性基于特定异构体在某些条件下和卤素原子的贡献。对新关联的交叉验证对应于留一法（Q ² _LOO）的确定系数）和遗漏（Q ² _LMO）相同且等于0.895，接近训练集的确定系数（R ² = 0.900）。进一步的统计参数还确认了新模型的拟合优度，预测优度，准确性和精确度。对于11种化合物，这些化合物可从多重线性回归分析（MLR）和非线性多重回归分析（MNLR）QSAR模型获得拟精神活性的外部验证数据集的输出，则新化合物的均方根误差（RMSE）值在模型中，MLR和MNLR分别为0.172、0.624和0.539。