A Novel Mixture Model to Estimate the Time to Drug Effect Onset and Its Association with Covariates

发布者:文明办发布时间:2020-11-19浏览次数:209


主讲人:姚音  复旦大学教授


时间:2020年11月20日14:00


地点:3号楼301室


举办单位:数理学院


主讲人介绍:姚音,教授,博士生导师。哥伦比亚大学遗传学博士,法国里昂国际癌症研究所博士后。复旦大学现代人类学研究中心特聘教授,美国NIH精神疾病研究所研究员。多年来一致从事复杂疾病遗传易感性研究。先后主持和承担了多个NIH课题,具有丰富的复杂疾病遗传易感性研究关联分析工作经验。在Nature  Genetics、Science、American Journal of Human Genetics、Cancer  Research等国际学术刊物发表论文100余篇。


内容介绍:Drugs take effect at different times in different individuals. Consequently,  researchers seek to examine how the timing of the biological response to drugs  may be affected by factors such as gender, genotypes, age, or baseline symptom  scores. Methods: Typically, studies measure symptoms immediately after the  initiation of drug treatment and then at a sequence of later time points. In  this study, we develop a statistical mixture model for analyzing such  longitudinal data. Our method estimates the onset of drug effect and assesses  the association between the probability distribution of the onset times and  possible contributing factors. Our mixture model treats the timing of onset as  missing for each individual but restricts it, for simplicity, to two possible  onset points, early or late. To estimate the model, we use an  expectation-maximization-based approach and provide the general formulas of the  variance and covariance matrix for the estimated parameters. Results: We  evaluate the model’s overall utility and performance via simulation studies. In  addition, we illustrate its use by application to longitudinal data from the  Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study. The  algorithm identified age and anxiety status as significant factors in affecting  the onset distribution of citalopram (Celexa).