Predicting human oral bioavailability of a compound: Development of a novel quantitative structure-bioavailability relationship
Predicting human oral bioavailability of a compound: Development of a novel quantitative structure-bioavailability relationship
复制标题
DOI：
10.1023/a:1007556711109
复制
发表时间：
2000-06-01
影响因子：
3.7
通讯作者：
Yu, LX
中科院分区：
文献类型：
--
作者：
Andrews, CW;Bennett, L;Yu, LX
目的。本研究的目的是开发用于药物发现和开发的定量结构-生物利用度关系 (QSBR) 模型。方法。具有人体口服生物利用度的药物数据库以电子形式组装，结构采用 SMILES 格式。使用该数据库，使用逐步回归程序将人类的口服生物利用度与药物的亚结构片段联系起来。将回归模型与 Lipinski 的五法则进行比较。结果。人类口服生物利用度数据库包含 591 种化合物。建立了采用 85 个描述符的回归模型，根据化合物的分子结构预测其人体口服生物利用度。与 Lipinski 的五法则相比，假阴性预测从 5% 减少到 3%，假阳性预测从 78% 减少到 53%。确定了一组子结构描述符以显示哪些片段倾向于增加/减少人类口服生物利用度。结论。开发了一种新的定量结构-生物利用度关系（QSBR）。尽管存在很大程度的实验误差，但该模型具有合理的预测性并且经得起交叉验证。与 Lipinski 的五法则相比，QSBR 模型能够减少误报预测。
Purpose. The purpose of this investigation was to develop a quantitative structure-bioavailability relationship (QSBR) model for drug discovery and development.Methods. A database of drugs with human oral bioavailability was assembled in electronic form with structure in SMILES format. Using that database, a stepwise regression procedure was used to link oral bioavailability in humans and substructural fragments in drugs. The regression model was compared with Lipinski's Rule of Five.Results. The human oral bioavailability database contains 591 compounds. A regression model employing 85 descriptors was built to predict the human oral bioavailability of a compound based on its molecular structure. Compared to Lipinski's Rule of Five, the false negative predictions were reduced from 5% to 3% while the false positive predictions decreased from 78% to 53%. A set of substructural descriptors was identified to show which fragments tend to increase/ decrease human oral bioavailability.Conclusions. A novel quantitative structure-bioavailability relationship (QSBR) was developed. Despite a large degree of experimental error, the model was reasonably predictive and stood up to crossvalidation. When compared to Lipinski's Rule of Five, the QSBR model was able to reduce false positive predictions.