Susceptibility gene identification and risk evaluation model construction by transcriptome-wide association analysis for salt sensitivity of blood pressure
Han Qi,
Yun-Yi Xie,
Xiao-Jun Yang,
Juan Xia,
Kuo Liu,
Feng-Xu Zhang,
Wen-Juan Peng,
Fu-Yuan Wen,
Bing-Xiao Li,
Bo-Wen Zhang,
Xin-Yue Yao,
Bo-Ya Li,
Hong-Dao Meng,
Zu-Min Shi,
Yang Wang,
Ling Zhang
Affiliations
Han Qi
Department of Epidemiology and Health Statistics, School of Public Health, Beijing Municipal Key Laboratory of Clinical Epidemiology, Capital Medical University
Yun-Yi Xie
Department of Epidemiology and Health Statistics, School of Public Health, Beijing Municipal Key Laboratory of Clinical Epidemiology, Capital Medical University
Xiao-Jun Yang
Department of Epidemiology and Health Statistics, School of Public Health, Beijing Municipal Key Laboratory of Clinical Epidemiology, Capital Medical University
Juan Xia
Department of Epidemiology and Health Statistics, School of Public Health, Beijing Municipal Key Laboratory of Clinical Epidemiology, Capital Medical University
Kuo Liu
Department of Epidemiology and Health Statistics, School of Public Health, Beijing Municipal Key Laboratory of Clinical Epidemiology, Capital Medical University
Feng-Xu Zhang
Department of Epidemiology and Health Statistics, School of Public Health, Beijing Municipal Key Laboratory of Clinical Epidemiology, Capital Medical University
Wen-Juan Peng
Department of Epidemiology and Health Statistics, School of Public Health, Beijing Municipal Key Laboratory of Clinical Epidemiology, Capital Medical University
Fu-Yuan Wen
Department of Epidemiology and Health Statistics, School of Public Health, Beijing Municipal Key Laboratory of Clinical Epidemiology, Capital Medical University
Bing-Xiao Li
Department of Epidemiology and Health Statistics, School of Public Health, Beijing Municipal Key Laboratory of Clinical Epidemiology, Capital Medical University
Bo-Wen Zhang
Department of Epidemiology and Health Statistics, School of Public Health, Beijing Municipal Key Laboratory of Clinical Epidemiology, Capital Medical University
Xin-Yue Yao
Department of Epidemiology and Health Statistics, School of Public Health, Beijing Municipal Key Laboratory of Clinical Epidemiology, Capital Medical University
Bo-Ya Li
Department of Epidemiology and Health Statistics, School of Public Health, Beijing Municipal Key Laboratory of Clinical Epidemiology, Capital Medical University
Hong-Dao Meng
Faculty of Information Technology, Beijing University of Technology
Zu-Min Shi
Human Nutrition Department, College of Health Sciences, QU Health, Qatar University
Yang Wang
Department of Cardiovascular Medicine, First Affiliated Hospital of Xi’an Jiaotong University
Ling Zhang
Department of Epidemiology and Health Statistics, School of Public Health, Beijing Municipal Key Laboratory of Clinical Epidemiology, Capital Medical University
Abstract Background Salt sensitivity of blood pressure (SSBP) is an intermediate phenotype of hypertension and is a predictor of long-term cardiovascular events and death. However, the genetic structures of SSBP are uncertain, and it is difficult to precisely diagnose SSBP in population. So, we aimed to identify genes related to susceptibility to the SSBP, construct a risk evaluation model, and explore the potential functions of these genes. Methods and results A genome-wide association study of the systemic epidemiology of salt sensitivity (EpiSS) cohort was performed to obtain summary statistics for SSBP. Then, we conducted a transcriptome-wide association study (TWAS) of 12 tissues using FUSION software to predict the genes associated with SSBP and verified the genes with an mRNA microarray. The potential roles of the genes were explored. Risk evaluation models of SSBP were constructed based on the serial P value thresholds of polygenetic risk scores (PRSs), polygenic transcriptome risk scores (PTRSs) and their combinations of the identified genes and genetic variants from the TWAS. The TWAS revealed that 2605 genes were significantly associated with SSBP. Among these genes, 69 were differentially expressed according to the microarray analysis. The functional analysis showed that the genes identified in the TWAS were enriched in metabolic process pathways. The PRSs were correlated with PTRSs in the heart atrial appendage, adrenal gland, EBV-transformed lymphocytes, pituitary, artery coronary, artery tibial and whole blood. Multiple logistic regression models revealed that a PRS of P < 0.05 had the best predictive ability compared with other PRSs and PTRSs. The combinations of PRSs and PTRSs did not significantly increase the prediction accuracy of SSBP in the training and validation datasets. Conclusions Several known and novel susceptibility genes for SSBP were identified via multitissue TWAS analysis. The risk evaluation model constructed with the PRS of susceptibility genes showed better diagnostic performance than the transcript levels, which could be applied to screen for SSBP high-risk individuals.
