中国麻风皮肤病杂志 ›› 2016, Vol. 32 ›› Issue (10): 581-583.

• 论著 • 上一篇    下一篇

基于易感基因的麻风风险预测模型的构建及效能研究

王真真1，刘红1,2，张福仁1,2   

  1. 1.山东省皮肤病性病防治研究所，山东省医学科学院，济南，250022；2.山东省皮肤病医院，山东大学，济南，250000
  • 出版日期:2016-10-15 发布日期:2018-12-17
  • 通讯作者: 张福仁，E-mail:zhangfuren@hotmail.com

Risk prediction model of leprosy based on the susceptibility genes and its power study

WANG Zhenzhen1, LIU Hong1,2, ZHANG Furen1,2   

  1. 1. Shandong Provincial Institute of Dermatology and Venereology, Academy of Medical Sciences, Jinan 250022, China;2.Shandong Provincial Dermatology Hospital, Shandong University, Jinan 250000, China;
  • Online:2016-10-15 Published:2018-12-17
  • Contact: ZHANG Furen, E-mail:zhangfuren@hotmail.com

PDF (PC)

210

摘要： 目的：基于麻风全基因组关联研究（GWAS）发现的18个易感位点，联合流行区域及家族史信息，建立中国人群麻风风险预测模型。方法：通过SNP位点等权重风险评分(GRS)和加权遗传风险评分(wGRS)法对5792例麻风患者和8097名正常对照计算易感位点的联合效应。以不同的方式联合遗传风险评分、地域及家族史信息，构建麻风风险预测模型，并绘制受试者工作曲线来评价模型。结果：wGRS联合地域及家族史信息建立的风险预测模型最佳，其曲线下面积（AUC）为0.758（95% CI：0.750~0.766）。结论：麻风易感位点间存在显著的联合作用，加权遗传风险评分联合地域及家族史信息建立的风险预测模型，能更好地预测麻风发病风险。

关键词: 麻风, 风险预测模型, 全基因组关联分析

Abstract: Objective: To evaluate the risk of leprosy by considering information on epidemic region and family history when combined with those? from 18 known susceptibility loci identified by genome-wide association studies (GWASs) associated with leprosy. Methods: Genetic risk score (GRS) and weighted genetic risk score (wGRS) were calculated to evaluate the joint effects of 18 susceptibility loci. Multiple models combining genetic loci and region and family history information were established. Receiver operating characteristic curve analysis was used to compare the power of different predictive models. Results: The model incorporating wGRS and information on epidemic region and family history was the best one to predict leprosy risk in Chinese population, with all area under curve of 0.758(95%CI: 0.750~0.766). Conclusion: Eighteen known susceptibility loci identified by GWASs jointly influence the leprosy risk?. The combination of 18 known susceptibility loci and information on epidemic region and family history can improve the performance of risk predictive model for the occurrence of leprosy.

Key words: leprosy, risk prediction model, GWAS