A persistent infection prolongs treatment duration and also enhances the chance of resistance development against antibiotics. Recently, a class of amphiphilic indole derivatives was discovered exhibiting bactericidal activity against both growing and nongrowing Mycobacterium bovis BCG (M. bovis BCG). These antibacterials are suggested to disturb the integrity and functioning of the cell membrane, a property that can help eradicate persistent organisms. This study article describes field‐based three‐dimensional quantitative structure–activity relationship (3D‐QSAR) studies of 79 amphiphilic indole derivatives. The aim of this QSAR study is to optimize this class of compounds for the development of more potent antimycobacterial agents. The results obtained indicate that steric interactions are crucial for antimycobacterial activity, while hydrogen bond donor groups participate negligibly in activity. The derived 3D‐QSAR models showed acceptable r² (0.91) and q² (0.91) with a root mean squared error (RMSE) of 0.08. The models were cross‐validated using the leave‐one‐out method. Applying the same QSAR model to another congeneric series of amphiphilic indoles externally validated the QSAR model. The model could appreciably predict the activity (pMIC₅₀) of this congeneric series of amphiphilic indoles, with an RMSE of 0.49, indicating the robustness of the model and its efficiency in predicting the potentially active compounds.

中文翻译：

---

3D QSAR研究两亲吲哚的抗分枝杆菌活性

持续性感染延长了治疗时间，也增加了对抗生素产生耐药性的机会。最近，发现了一类两亲的吲哚衍生物对生长和不生长的牛分枝杆菌BCG（M. bovis）均具有杀菌活性。卡介苗）。建议这些抗菌素干扰细胞膜的完整性和功能，这种特性可以帮助根除持久性有机体。本文研究了79种两亲吲哚衍生物的基于场的三维定量结构-活性关系（3D-QSAR）研究。这项QSAR研究的目的是优化此类化合物，以开发更有效的抗分枝杆菌药。获得的结果表明，空间相互作用对于抗分枝杆菌的活性至关重要，而氢键供体基团的参与微不足道。派生的3D-QSAR模型显示出可接受的r ²（0.91）和q ²（0.91），均方根误差（RMSE）为0.08。使用留一法对模型进行交叉验证。将相同的QSAR模型应用于另一个同类两亲性吲哚系列，从外部验证了QSAR模型。该模型可以明显预测该同类两亲吲哚系列的活性（pMIC ₅₀），RMSE为0.49，表明该模型的稳健性及其预测潜在活性化合物的效率。