Importance of GWAS risk loci and clinical data in predicting asthma using machine-learning approaches.

INTRODUCTION: To understand the risk factors of asthma, we combined genome-wide association study (GWAS) risk loci and clinical data in predicting asthma using machine-learning approaches. METHODS: A case-control study with 123 asthmatics and 100 controls was conducted in the Zhuang population in Guangxi. GWAS risk loci were detected using polymerase chain reaction, and clinical data were collected. Machine-learning approaches were used to identify the major factors that contribute to asthma. RESULTS: A total of 14 GWAS risk loci with clinical data were analyzed on the basis of 10 times the 10-fold cross-validation for all machine-learning models. Using GWAS risk loci or clinical data, the best performances exhibited area under the curve (AUC) values of 64.3% and 71.4%, respectively. Combining GWAS risk loci and clinical data, the XGBoost established the best model with an AUC of 79.7%, indicating that the combination of genetics and clinical data can enable improved performance. We then sorted the importance of features and found the top six risk factors for predicting asthma to be rs3117098, rs7775228, family history, rs2305480, rs4833095, and body mass index. CONCLUSION: Asthma-prediction models based on GWAS risk loci and clinical data can accurately predict asthma, and thus provide insights into the disease pathogenesis.

Association of GWAS-supported noncoding area loci rs404860, rs3117098, and rs7775228 with asthma in Chinese Zhuang population.

BACKGROUND: Asthma is a complicated and polygenic inheritance disease, and its prevalence increases worldwide. Recent genome-wide association studies (GWASs) identified a significant association of single nucleotide polymorphism with asthma in the Japanese population. This study aimed to examine the association of GWAS-supported noncoding area loci, namely rs404860, rs3117098, and rs7775228, with asthma in Chinese Zhuang population. METHODS: A case-control study involving 223 individuals, comprising 123 patients with asthma and 100 healthy controls, was conducted. Genotypes were determined by polymerase chain reaction (PCR)/ligase detection reaction assay. The association between gene polymorphisms and asthma risk was calculated by logistic regression analysis using different genetic models through comparisons of alleles (A vs a), homozygote genotypes (AA vs aa), heterozygote genotypes (Aa vs aa), dominant models (AA+Aa vs aa), and recessive models (AA vs. Aa+aa). RESULTS: The distribution of the genotype frequency of rs3117098 was statistically different between the case and control groups. For rs3117098, significant associations were observed through comparisons of alleles (OR: 1.832, 95% CI: 1.048-3.204, P = .034) and dominant models (OR: 2.065, 95% CI: 1.001-4.260, P = .050). The statistical analysis showed no significant difference for loci rs404860 and rs7775228 between patients with asthma and controls. CONCLUSION: rs3117098 may be the risk factor for asthma in Chinese Zhuang population.

The rs11191580 variant of the NT5C2 gene is associated with schizophrenia and symptom severity in a South Chinese Han population: evidence from GWAS

Objective: Recent genome-wide association studies have identified a significant relationship between the NT5C2 variant rs11191580 and schizophrenia (SCZ) in European populations. This study aimed to validate the association of rs11191580 polymorphism with SCZ risk in a South Chinese Han population. The relationship of this polymorphism with the severity of SCZ clinical symptoms was also explored. Methods: A case-control study was performed in 462 patients with SCZ and 598 healthy controls. Rs11191580 was genotyped by the Sequenom MassARRAY iPLEX platform. A total of 459 SCZ patients completed the Positive and Negative Syndrome Scale (PANSS) evaluation. Data were analyzed by PLINK software. Results: We confirmed an association of the rs11191580 polymorphism with SCZ risk in South Chinese Han under a dominant genetic model (ORadj = 0.769; 95%CIadj = 0.600-0.984; padj = 0.037). PANSS scores showed a significant association between variant rs11191580 and total score (padj = 0.032), lack of response scale score (padj = 0.022), and negative scale score (additive: padj = 0.004; dominant: padj = 0.016; recessive: padj = 0.021) after data were adjusted for age and sex. Conclusion: NT5C2 variant rs11191580 conferred susceptibility to SCZ and affected the clinical symptoms of SCZ in a South Chinese Han population.