摘要
在抗乳腺癌药物研发中,为了节省时间成本,建立化合物预测模型来筛选活性化合物是一种有效的方法。本文根据提供的乳腺癌治疗靶标雌激素受体α亚型(Estrogen receptors alpha, ERα)拮抗剂信息,利用Lasso回归与随机森林相结合的方法,对数据降维并筛选出影响生物活性的主要变量,明确了模型建立的方向;在此基础上,建立了关于生物活性的随机森林回归模型并进行了预测。
In the research and development of anti-breast cancer drugs, it is an effective method to establish a compound prediction model to screen active compounds in order to save time and cost. According to the information of Estrogen receptors alpha (ERα) antagonist, which is the target of breast cancer treatment, this paper uses Lasso regression and random forest to reduce the dimension of the data and screen out the main variables that affect the biological activity, and makes clear the direction of establishing the model. On this basis, a random forest regression model about biological activity was established and predicted.
出处
《建模与仿真》
2023年第2期1583-1592,共10页
Modeling and Simulation
