MTML: An Efficient Multitrait Multilocus GWAS Method Based on the Cauchy Combination Test.

Genome-wide association study (GWAS) by measuring the joint effect of multiple loci on multiple traits, has recently attracted interest, due to the decreased costs of high-throughput genotyping and phenotyping technologies. Previous studies mainly focused on either multilocus models that identify associations with a single trait or multitrait models that scan a single marker at a time. Since these types of models cannot fully utilize the association information, the powers of the tests are usually low. To potentially address this problem, we present here a multitrait multilocus (MTML) modeling framework that implements in three steps: (1) simplify the complex calculation; (2) reduce the model dimension; (3) integrate the joint contribution of single markers to multiple traits by Cauchy combination. The performances of MTML are evaluated and compared with other three published methods by Monte Carlo simulations. Simulation results show that MTML is more powerful for quantitative trait nucleotide detection and robust for various numbers of traits. In the meanwhile, MTML can effectively control type I error rate at a reasonable level. Real data analysis of Arabidopsis thaliana shows that MTML identifies more pleiotropic genetic associations. Therefore, we conclude that MTML is an efficient GWAS method for joint analysis of multiple quantitative traits. The R package MTML, which facilitates the implementation of the proposed method, is publicly available on GitHub https://github.com/Guohongping/MTML.

Pleiotropic genetic association analysis with multiple phenotypes using multivariate response best-subset selection.

Genetic pleiotropy refers to the simultaneous association of a gene with multiple phenotypes. It is widely distributed in the whole genome and can help to understand the common genetic mechanism of diseases or traits. In this study, a multivariate response best-subset selection (MRBSS) model based pleiotropic association analysis method is proposed. Different from the traditional genetic association model, the high-dimensional genotypic data are viewed as response variables while the multiple phenotypic data as predictor variables. Moreover, the response best-subset selection procedure is converted into an 0-1 integer optimization problem by introducing a separation parameter and a tuning parameter. Furthermore, the model parameters are estimated by using the curve search under the modified Bayesian information criterion. Simulation experiments show that the proposed method MRBSS remarkably reduces the computational time, obtains higher statistical power under most of the considered scenarios, and controls the type I error rate at a low level. The application studies in the datasets of maize yield traits and pig lipid traits further verifies the effectiveness.

A Two-Stage Mutual Information Based Bayesian Lasso Algorithm for Multi-Locus Genome-Wide Association Studies.

Genome-wide association study (GWAS) has turned out to be an essential technology for exploring the genetic mechanism of complex traits. To reduce the complexity of computation, it is well accepted to remove unrelated single nucleotide polymorphisms (SNPs) before GWAS, e.g., by using iterative sure independence screening expectation-maximization Bayesian Lasso (ISIS EM-BLASSO) method. In this work, a modified version of ISIS EM-BLASSO is proposed, which reduces the number of SNPs by a screening methodology based on Pearson correlation and mutual information, then estimates the effects via EM-Bayesian Lasso (EM-BLASSO), and finally detects the true quantitative trait nucleotides (QTNs) through likelihood ratio test. We call our method a two-stage mutual information based Bayesian Lasso (MBLASSO). Under three simulation scenarios, MBLASSO improves the statistical power and retains the higher effect estimation accuracy when comparing with three other algorithms. Moreover, MBLASSO performs best on model fitting, the accuracy of detected associations is the highest, and 21 genes can only be detected by MBLASSO in Arabidopsis thaliana datasets.

Malaria from hyperendemicity to elimination in Hekou County on China-Vietnam border: an ecological study.

BACKGROUND: Malaria control and elimination are challenged by diversity and complexity of the determinants on the international border in the Great Mekong Sub-region. Hekou, a Chinese county on the China-Vietnam border, was used to document Chinese experiences and lessons for malaria control and elimination. METHODS: The design was an ecological study. Malaria burden before 1951 and procedures of 64 years (1952-2015) from malaria hyperendemicity to elimination are described. Single and bilinear regression analysis was utilized to analyse the relationship between the annual malaria incidence (AMI) and gross domestic product (GDP), urbanization rate, and banana planting area (BPA). RESULTS: There was a huge malaria burden before 1951. AMI was reduced from 358.62 per 1000 person-years in 1953 to 5.69 per 1000 person-years in 1960. A system of primary health services, comprising three levels of county township hospitals and village health stations maintained malaria control and surveillance activities in changing political and social-economic settings. However, potential under-reported of malaria and market-oriented healthcare led to a malaria epidemic in 1987. Strong political commitment reoriented malaria from a control to an elimination programme. High coverage of malaria intervention and population access to intervention was crucial for malaria control and elimination; meanwhile, AMI was closely associated with socio-economic development, correlation coefficients (R) -0.6845 (95% CI -0.7978, -0.6845) for national GDP, -0.7014 (-0.8093, -0.7014) for national urbanization rate and -0.5563 (-0.7147, -0.3437) for BPA. CONCLUSIONS: Multifactor, including political commitment, effective interventions, social and economic development and changing ecological environment, and the complicated interactions between these factors contribute to malaria elimination in Hekou County.

Subclassification of lung adenocarcinoma through comprehensive multi-omics data to benefit survival outcomes.

OBJECTIVES: Lung adenocarcinoma (LUAD) is the most common subtype of non-small cell lung cancer. Understanding the molecular mechanisms underlying tumor progression is of great clinical significance. This study aims to identify novel molecular markers associated with LUAD subtypes, with the goal of improving the precision of LUAD subtype classification. Additionally, optimization efforts are directed towards enhancing insights from the perspective of patient survival analysis. MATERIALS AND METHODS: We propose an innovative feature-selection approach that focuses on LUAD classification, which is comprehensive and robust. The proposed method integrates multi-omics data from The Cancer Genome Atlas (TCGA) and leverages a synergistic combination of max-relevance and min-redundancy, least absolute shrinkage and selection operator, and Boruta algorithms. These selected features were deployed in six machine-learning classifiers: logistic regression, random forest, support vector machine, naive Bayes, k-Nearest Neighbor, and XGBoost. RESULTS: The proposed approach achieved an area under the receiver operating characteristic curve (AUC) of 0.9958 for LR. Notably, the accuracy and AUC of a composite model incorporating copy number, methylation, as well as RNA- sequencing data for expression of exons, genes, and miRNA mature strands surpassed the accuracy and AUC metrics of models with single-omics data or other multi-omics combinations. Survival analyses, revealed the SVM classifier to elicit optimal classification, outperforming that achieved by TCGA. To enhance model interpretability, SHapley Additive exPlanations (SHAP) values were utilized to elucidate the impact of each feature on the predictions. Gene Ontology (GO) enrichment analysis identified significant biological processes, molecular functions, and cellular components associated with LUAD subtypes. CONCLUSION: In summary, our feature selection process, based on TCGA multi-omics data and combined with multiple machine learning classifiers, proficiently identifies molecular subtypes of lung adenocarcinoma and their corresponding significant genes. Our method could enhance the early detection and diagnosis of LUAD, expedite the development of targeted therapies and, ultimately, lengthen patient survival.

Screening properties of trend tests in genetic association studies.

In genome-wide association study, extracting disease-associated genetic variants among millions of single nucleotide polymorphisms is of great importance. When the response is a binary variable, the Cochran-Armitage trend tests and associated MAX test are among the most widely used methods for association analysis. However, the theoretical guarantees for applying these methods to variable screening have not been built. To fill this gap, we propose screening procedures based on adjusted versions of these methods and prove their sure screening properties and ranking consistency properties. Extensive simulations are conducted to compare the performances of different screening procedures and demonstrate the robustness and efficiency of MAX test-based screening procedure. A case study on a dataset of type 1 diabetes further verifies their effectiveness.

Divided-and-combined omnibus test for genetic association analysis with high-dimensional data.

Advances in biologic technology enable researchers to obtain a huge amount of genetic and genomic data, whose dimensions are often quite high on both phenotypes and variants. Testing their association with multiple phenotypes has been a hot topic in recent years. Traditional single phenotype multiple variant analysis has to be adjusted for multiple testing and thus suffers from substantial power loss due to ignorance of correlation across phenotypes. Similarity-based method, which uses the trace of product of two similarity matrices as a test statistic, has emerged as a useful tool to handle this problem. However, it loses power when the correlation strength within multiple phenotypes is middle or strong, for some signals represented by the eigenvalues of phenotypic similarity matrix are masked by others. We propose a divided-and-combined omnibus test to handle this drawback of the similarity-based method. Based on the divided-and-combined strategy, we first divide signals into two groups in a series of cut points according to eigenvalues of the phenotypic similarity matrix and combine analysis results via the Cauchy-combined method to reach a final statistic. Extensive simulations and application to a pig data demonstrate that the proposed statistic is much more powerful and robust than the original test under most of the considered scenarios, and sometimes the power increase can be more than 0.6. Divided-and-combined omnibus test facilitates genetic association analysis with high-dimensional data and achieves much higher power than the existing similarity based method. In fact, divided-and-combined omnibus test can be used whenever the association analysis between two multivariate variables needs to be conducted.

A genome-wide cross-cancer meta-analysis highlights the shared genetic links of five solid cancers.

Breast, ovarian, prostate, lung, and head/neck cancers are five solid cancers with complex interrelationships. However, the shared genetic factors of the five cancers were often revealed either by the combination of individual genome-wide association study (GWAS) approach or by the fixed-effect model-based meta-analysis approach with practically impossible assumptions. Here, we presented a random-effect model-based cross-cancer meta-analysis framework for identifying the genetic variants jointly influencing the five solid cancers. A comprehensive genetic correlation analysis (genome-wide, partitioned, and local) approach was performed by using GWAS summary statistics of the five cancers, and we observed three cancer pairs with significant genetic correlation: breast-ovarian cancer (r g = 0.221, p = 0.0003), breast-lung cancer (r g = 0.234, p = 7.6 × 10-6), and lung-head/neck cancer (r g = 0.652, p = 0.010). Furthermore, a random-effect model-based cross-trait meta-analysis was conducted for each significant cancer pair, and we found 27 shared genetic loci between breast and ovarian cancers, 18 loci between breast and lung cancers, and three loci between lung and head/neck cancers. Functional analysis indicates that the shared genes are enriched in human T-cell leukemia virus 1 infection (HTLV-1) and antigen processing and presentation (APP) pathways. Our study investigates the shared genetic links across five solid cancers and will help to reveal their potential molecular mechanisms.

Identifying shared genetic loci between coronavirus disease 2019 and cardiovascular diseases based on cross-trait meta-analysis.

People with coronavirus disease 2019 (COVID-19) have different mortality or severity, and this clinical outcome is thought to be mainly attributed to comorbid cardiovascular diseases. However, genetic loci jointly influencing COVID-19 and cardiovascular disorders remain largely unknown. To identify shared genetic loci between COVID-19 and cardiac traits, we conducted a genome-wide cross-trait meta-analysis. Firstly, from eight cardiovascular disorders, we found positive genetic correlations between COVID-19 and coronary artery disease (CAD, R g = 0.4075, P = 0.0031), type 2 diabetes (T2D, R g = 0.2320, P = 0.0043), obesity (OBE, R g = 0.3451, P = 0.0061), as well as hypertension (HTN, R g = 0.233, P = 0.0026). Secondly, we detected 10 shared genetic loci between COVID-19 and CAD, 3 loci between COVID-19 and T2D, 5 loci between COVID-19 and OBE, and 21 loci between COVID-19 and HTN, respectively. These shared genetic loci were enriched in signaling pathways and secretion pathways. In addition, Mendelian randomization analysis revealed significant causal effect of COVID-19 on CAD, OBE and HTN. Our results have revealed the genetic architecture shared by COVID-19 and CVD, and will help to shed light on the molecular mechanisms underlying the associations between COVID-19 and cardiac traits.

A maximum-type microbial differential abundance test with application to high-dimensional microbiome data analyses.

Background: High-throughput metagenomic sequencing technologies have shown prominent advantages over traditional pathogen detection methods, bringing great potential in clinical pathogen diagnosis and treatment of infectious diseases. Nevertheless, how to accurately detect the difference in microbiome profiles between treatment or disease conditions remains computationally challenging. Results: In this study, we propose a novel test for identifying the difference between two high-dimensional microbiome abundance data matrices based on the centered log-ratio transformation of the microbiome compositions. The test p-value can be calculated directly with a closed-form solution from the derived asymptotic null distribution. We also investigate the asymptotic statistical power against sparse alternatives that are typically encountered in microbiome studies. The proposed test is maximum-type equal-covariance-assumption-free (MECAF), making it widely applicable to studies that compare microbiome compositions between conditions. Our simulation studies demonstrated that the proposed MECAF test achieves more desirable power than competing methods while having the type I error rate well controlled under various scenarios. The usefulness of the proposed test is further illustrated with two real microbiome data analyses. The source code of the proposed method is freely available at https://github.com/Jiyuan-NYU-Langone/MECAF. Conclusions: MECAF is a flexible differential abundance test and achieves statistical efficiency in analyzing high-throughput microbiome data. The proposed new method will allow us to efficiently discover shifts in microbiome abundances between disease and treatment conditions, broadening our understanding of the disease and ultimately improving clinical diagnosis and treatment.

Duration of immunogenicity of high-dose and prolonged-schedule hepatitis B vaccine among patients with chronic kidney disease: A one year follow-up study in China.

BACKGROUND: Patients with chronic kidney disease (CKD) have immunological defects that result in reduced production and faster decay of anti-HBs after hepatitis B vaccination. We assessed the duration of the immunogenicity after four-standard-dose and four-triple-dose regimens among patients with CKD. METHODS: A randomized controlled trial was conducted between May 2019 and February 2020. Patients were randomly allocated to receive three or four doses of 20 µg    , or four doses of 60 µg   of hepatitis B vaccine. Immunogenicity was assessed for 18 months till February 2021. RESULTS: Between months 7 and 18, the seroconversion rate decreased from 81.7% (58/71) to 64.3% (36/56) in IM20 × 3 group, from 93.0% (66/71) to 77.4% (41/53) in IM20 × 4 group, and from 93.2% (68/73) to 90.7% (49/54) in IM60 × 4 group. Seroconversion was higher in IM60 × 4 group than in IM20 × 3 group at month 18 (P < 0.05). In multivariate analysis, CKD patients without immune suppression or hormone therapy or patients with IM60 × 4 were more likely to have durable immunogenicity at month 18. CONCLUSIONS: Patients receiving four-triple-dose regimen of hepatitis B vaccine showed improved duration of immunogenicity at the one-year follow-up. CLINICAL TRIAL REGISTRATION: ClinicalTrials.gov (NCT03962881).

Immunogenicity and safety of a high-dose and prolonged-schedule hepatitis B vaccine among chronic kidney disease patients: a randomized, parallel-controlled trial.

Background: The immunogenicity against hepatitis B vaccine is unsatisfactory in the chronic kidney disease (CKD) patients, and studies evaluating augmented vaccine regimens to enhance immunogenicity have been inconclusive.Objectives: To evaluate the immunogenicity and safety of four-standard-dose and four-triple-dose regimens hepatitis B vaccine among CKD patients in China.Research design and methods: We conducted a multicenter, randomized, parallel-controlled trial including 273 patients with CKD who were randomly allocated to receive 3 or 4 doses of 20 or 60 µg of recombinant hepatitis B vaccine.Main outcome measures: Seroconversion rates, high-level response rates, and geometric mean concentrations (GMCs) of anti-HBs at months 3 and 7.Results: The seroconversion rates and high-level responses in the IM20 × 4 group and the IM60 × 4 group were higher than those in the IM20 × 3 group at months 3 and 7 (P < 0.05). The IM60 × 4 group had better immune responses than the IM20 × 4 group at month 3 (P < 0.05); however, no significant difference was noted at month 7 (P > 0.05).Conclusions: Both the four-standard-dose and four-triple-dose regimens improved immune response compared to the three-standard-dose regimen of hepatitis B vaccination in CKD patients, and the additional effect of higher dose was minimal.Trial registration: The trial is registered at ClinicalTrials.gov (CT.gov identifier: NCT03962881).

Predicting related factors of immunological response to hepatitis B vaccine in hemodialysis patients based on integration of decision tree classification and logistic regression.

The non/hypo-response rate of the hepatitis B vaccine among hemodialysis (HD) patients is still high, it is of great significance to explore the influencing factors and their relationships. To study the related factors and their relationships using logistic regression model and Chi-squared Automatic Interaction Detection (CHAID) decision tree model. A randomized controlled trial was conducted between February 2014 and May 2015 in China. HD patients being serologically negative for HBsAg and anti-HBs were randomly assigned to receive three intramuscular injections of the standard dose (20 µg) or high dose (60 µg) of recombinant hepatitis B vaccine at months 0, 1, and 6. Those with anti-HBs concentrations <100 mIU/mL, and ≥100 mIU/mL at month 7 were considered as non/hypo-response and high-level response, respectively. The non/hypo-response was 31.34% (89/284). After adjustment for confounders, logistic analysis showed that males (OR = 2.203, 95%CI: 1.109-4.367) and those with higher dialysis frequency (>4 times per 2 weeks) (OR = 1.918, 95%CI: 1.015-3.626) had a significant risk of non/hypo-response. While the CHAID analysis showed that gender, dose, and dialysis frequency were influencing factors of non/hypo-response, and gender is most important. The interaction between gender and dialysis frequency had the greatest effect on immunization, and followed by the interaction between dialysis frequency and vaccine dose. Taken together, gender, dose and dialysis frequency were influencing factors of hepatitis B vaccine in HD patients.

Long-term persistent immunogenicity after successful standard and triple-dosed hepatitis B vaccine in hemodialysis patients: A 3-year follow-up study in China.

BACKGROUND: Although the efficacy of hepatitis B vaccines among hemodialysis patients has been documented, the long-term persistence of immunogenicity in this population remains largely unknown. We explored the long-term persistence of immunogenicity induced by different hepatitis B vaccine regimens in hemodialysis patients. METHODS: In initial study, we conducted a randomized, multicenter, double-blind, parallel-controlled trial among hemodialysis patients in 13 hospitals in Shanxi Province, China. A total of 352 hemodialysis patients were allocated to receive 3-dose 20 µg (IM20 group) and 3-dose 60 µg (IM60 group) recombinant hepatitis B vaccine at months 0, 1, and 6. Vaccine-induced immune responses were measured at month 7. In this study, the responders (anti-HBs ≥ 10 mIU/mL) were followed up at months 18, 24, 30, 36 and 42, respectively. We used the generalized log-rank test and generalized estimating equations (GEE) to analyze the long-term durability of responses and the kinetics of anti-HBs levels, respectively. RESULTS: A total of 284 patients were involved in the extended follow-up period. The duration of vaccine-induced response with 75% of patients maintained protective antibody were 12 months and 18 months in the IM20 group and IM60 group, respectively (P = 0.291). The long-term persistent immunogenicity induced by 3-dose 60 µg was more satisfactory than that by 3-dose 20 µg hepatitis B vaccine in patients with hemodialysis duration ≥ five years (P = 0.023). The peak anti-HBs levels in 100-1000 mIU/mL or ≥ 1000 mIU/mL were more likely to maintain long-term protective antibody compared to anti-HBs levels in 10-100 mIU/mL (P ＜ 0.05). The kinetic profile was similar between the two groups (P = 0.334). CONCLUSION: High-dose 60 µg hepatitis B vaccine could lead a satisfactory long-term durability of immunogenicity among patients with hemodialysis duration of five years or more. Peak anti-HBs level after vaccination was associated with the long-term persistence of immunogenicity.

Identifying Shared Risk Genes for Asthma, Hay Fever, and Eczema by Multi-Trait and Multiomic Association Analyses.

Asthma, hay fever and eczema are three comorbid diseases with high prevalence and heritability. Their common genetic architectures have not been well-elucidated. In this study, we first conducted a linkage disequilibrium score regression analysis to confirm the strong genetic correlations between asthma, hay fever and eczema. We then integrated three distinct association analyses (metaCCA multi-trait association analysis, MAGMA genome-wide and MetaXcan transcriptome-wide gene-based tests) to identify shared risk genes based on the large-scale GWAS results in the GeneATLAS database. MetaCCA can detect pleiotropic genes associated with these three diseases jointly. MAGMA and MetaXcan were performed separately to identify candidate risk genes for each of the three diseases. We finally identified 150 shared risk genes, in which 60 genes are novel. Functional enrichment analysis revealed that the shared risk genes are enriched in inflammatory bowel disease, T cells differentiation and other related biological pathways. Our work may provide help on treatment of asthma, hay fever and eczema in clinical applications.

Systematic module approach identifies altered genes and pathways in four types of ovarian cancer.

The present study aimed to identify altered genes and pathways associated with four histotypes of ovarian cancer, according to the systematic tracking of dysregulated modules of reweighted protein­protein interaction (PPI) networks. Firstly, the PPI network and gene expression data were initially integrated to infer and reweight normal ovarian and four types of ovarian cancer (endometrioid, serous, mucinous and clear cell carcinoma) PPI networks based on Spearman's correlation coefficient. Secondly, modules in the PPI network were mined using a clique­merging algorithm and the differential modules were identified through maximum weight bipartite matching. Finally, the gene compositions in the altered modules were analyzed, and pathway functional enrichment analyses for disrupted module genes were performed. In five conditional­specific networks, universal alterations in gene correlations were revealed, which leads to the differential correlation density among disrupted module pairs. The analyses revealed 28, 133, 139 and 33 altered modules in endometrioid, serous, mucinous and clear cell carcinoma, respectively. Gene composition analyses of the disrupted modules revealed five common genes (mitogen­activated protein kinase 1, phosphoinositide 3­kinase­encoding catalytic 110­KDα, AKT serine/threonine kinase 1, cyclin D1 and tumor protein P53) across the four subtypes of ovarian cancer. In addition, pathway enrichment analysis confirmed one common pathway (pathways in cancer), in the four histotypes. This systematic module approach successfully identified altered genes and pathways in the four types of ovarian cancer. The extensive differences of gene correlations result in dysfunctional modules, and the coordinated disruption of these modules contributes to the development and progression of ovarian cancer.