计算机辅助药物代谢不良反应风险评价研究

The adverse reaction of drug metabolism is an important reason causing drug safety issues. Evaluation of the risk of adverse reaction of drug metabolism in early stage of drug research and development by using computer prediction models, can reduce the risk of drug development. However, the accuracy of current drug metabolite prediction models are not high enough, and the prediction of drug toxicity without considering the effect of drug metabolites, making it difficult to achieve this goal. In this project, prediction models for substrates, inhibitors and inducers of metabolic enzyme, and prediction models for substrates and inhibitorsof transporter will be built. Similarity comparison and statistical learning method will be used to construct the prediction model of drug metabolites. Then, the prediction models for toxic effects, such as liver toxicity, of drugs and their metabolites will be constructed. The pharmacokinetics experimentals will be used to test and optimize the prediction models. In this sdudy, the computer-aided risk evaluation system for adverse reaction of drug metabolism is anticipated to be built. Meanwhile, 4-6 papers will be published, and 1-2 patents or software copyrights will be applied.

药物代谢不良反应是导致药物安全性问题的重要原因。通过计算机预测模型在药物研发早期进行药物代谢不良反应风险评价，可以降低药物研发风险。但目前药物代谢物预测的准确性不够高，预测药物的毒副作用时没有考虑药物代谢物的影响，使得难以达成这一目标。本项目拟在前期研究的基础上，构建代谢酶底物、抑制剂和诱导剂预测模型，转运体底物、抑制剂预测模型，通过相似性比对和统计学习等方法来构建药物代谢物预测模型，构建药物及代谢物肝毒性等毒副作用预测模型。然后结合药代动力学试验，对预测模型进行检验和优化。本项目将构建计算机辅助药物代谢不良反应风险评价体系，发表论文4-6篇，申请专利、软件版权1-2项。

药物代谢不良反应是导致药物安全性问题的重要原因。通过计算机预测模型在药物研发早期进行药物代谢不良反应风险评价，可以降低药物研发风险。本项目在前期研究的基础上，构建了代谢酶抑制剂（底物）预测模型，转运体底物、抑制剂预测模型，药物代谢物及药物代谢位点预测模型，药物半衰期预测模型，药物及代谢物肝毒性预测模型。我们发现新的CYP450代谢途径在目标化合物体内代谢活化中的作用，发现目标化合物可影响主要药物代谢CYP450酶的体内基因和活性水平，发现新的CYP450代谢途径在目标化合物介导的肝毒性中具有潜在作用。发表论文4篇，申请专利1项。