Gene set analysis and network analysis for genome-wide association studies

Pathway-based analysis of genome-wide association study of circadian phenotypes

Sleepiness affects normal social life, which attracts more and more attention. Circadian phenotypes contribute to obvious individual differences in susceptibility to sleepiness. We aimed to identify candidate single nucleotide polymorphisms (SNPs) which may cause circadian phenotypes, elucidate the potential mechanisms, and generate corresponding SNP-gene-pathways. A genome-wide association studies (GWAS) dataset of circadian phenotypes was utilized in the study. Then, the Identify Candidate Causal SNPs and Pathways analysis was employed to the GWAS dataset after quality control filters. Furthermore, genotype-phenotype association analysis was performed with HapMap database. Four SNPs in three different genes were determined to correlate with usual weekday bedtime, totally providing seven hypothetical mechanisms. Eleven SNPs in six genes were identified to correlate with usual weekday sleep duration, which provided six hypothetical pathways. Our results demonstrated that fifteen candidate SNPs in eight genes played vital roles in six hypothetical pathways implicated in usual weekday bedtime and six potential pathways involved in usual weekday sleep duration.

Methods and Tools in Genome-wide Association Studies

Many traits, such as height, the response to a given drug, or the susceptibility to certain diseases are presumably co-determined by genetics. Especially in the field of medicine, it is of major interest to identify genetic aberrations that alter an individual's risk to develop a certain phenotypic trait. Addressing this question requires the availability of comprehensive, high-quality genetic datasets. The technological advancements and the decreasing cost of genotyping in the last decade led to an increase in such datasets. Parallel to and in line with this technological progress, an analysis framework under the name of genome-wide association studies was developed to properly collect and analyze these data. Genome-wide association studies aim at finding statistical dependencies-or associations-between a trait of interest and point-mutations in the DNA. The statistical models used to detect such associations are diverse, spanning the whole range from the frequentist to the Bayesian setting.Since genetic datasets are inherently high-dimensional, the search for associations poses not only a statistical but also a computational challenge. As a result, a variety of toolboxes and software packages have been developed, each implementing different statistical methods while using various optimizations and mathematical techniques to enhance the computations.This chapter is devoted to the discussion of widely used methods and tools in genome-wide association studies. We present the different statistical models and the assumptions on which they are based, explain peculiarities of the data that have to be accounted for and, most importantly, introduce commonly used tools and software packages for the different tasks in a genome-wide association study, complemented with examples for their application.

Genome-wide analysis identifies an african-specific variant in SEMA4D associated with body mass index

OBJECTIVE

The prevalence of obesity varies between ethnic groups. No genome-wide association study (GWAS) for body mass index (BMI) has been conducted in continental Africans.

METHODS

We performed a GWAS for BMI in 1,570 West Africans (WA). Replication was conducted in independent samples of WA (n = 1,411) and African Americans (AA) (n = 9,020).

RESULTS

We identified a novel genome-wide significant African-specific locus for BMI (SEMA4D, rs80068415; minor allele frequency = 0.008, P = 2.10 × 10-8 ). This finding was replicated in independent samples of WA (P = 0.013) and AA (P = 0.017). Individuals with obesity had higher serum SEMA4D levels compared to those without obesity (P < 0.0001), and elevated levels of serum SEMA4D were associated with increased obesity risk (OR = 4.2, P < 1 × 10-4 ). The prevalence of obesity was higher in individuals with the CT versus TT genotypes (55.6% vs. 22.9%).

CONCLUSIONS

A novel variant in SEMA4D was significantly associated with BMI. Carriers of the C allele were 4.6 BMI units heavier than carriers of the T allele (P = 0.0007). This variant is monomorphic in Europeans and Asians, highlighting the importance of studying diverse populations. While there is evidence for the involvement of SEMA4D in inflammatory processes, this study is the first to implicate SEMA4D in obesity pathophysiology.

A multistep approach to single nucleotide polymorphism-set analysis: an evaluation of power and type I error of gene-based tests of association after pathway-based association tests

The aggregation of functionally associated variants given a priori biological information can aid in the discovery of rare variants associated with complex diseases. Many methods exist that aggregate rare variants into a set and compute a single p value summarizing association between the set of rare variants and a phenotype of interest. These methods are often called gene-based, rare variant tests of association because the variants in the set are often all contained within the same gene. A reasonable extension of these approaches involves aggregating variants across an even larger set of variants (eg, all variants contained in genes within a pathway). Testing sets of variants such as pathways for association with a disease phenotype reduces multiple testing penalties, may increase power, and allows for straightforward biological interpretation. However, a significant variant-set association test does not indicate precisely which variants contained within that set are causal. Because pathways often contain many variants, it may be helpful to follow-up significant pathway tests by conducting gene-based tests on each gene in that pathway to narrow in on the region of causal variants. In this paper, we propose such a multistep approach for variant-set analysis that can also account for covariates and complex pedigree structure. We demonstrate this approach on simulated phenotypes from Genetic Analysis Workshop 19. We find generally better power for the multistep approach when compared to a more conventional, single-step approach that simply runs gene-based tests of association on each gene across the genome. Further work is necessary to evaluate the multistep approach on different data sets with different characteristics.

Genome-wide association study in obsessive-compulsive disorder: results from the OCGAS

Obsessive-compulsive disorder (OCD) is a psychiatric condition characterized by intrusive thoughts and urges and repetitive, intentional behaviors that cause significant distress and impair functioning. The OCD Collaborative Genetics Association Study (OCGAS) is comprised of comprehensively assessed OCD patients with an early age of OCD onset. After application of a stringent quality control protocol, a total of 1065 families (containing 1406 patients with OCD), combined with population-based samples (resulting in a total sample of 5061 individuals), were studied. An integrative analyses pipeline was utilized, involving association testing at single-nucleotide polymorphism (SNP) and gene levels (via a hybrid approach that allowed for combined analyses of the family- and population-based data). The smallest P-value was observed for a marker on chromosome 9 (near PTPRD, P=4.13 × 10(-)(7)). Pre-synaptic PTPRD promotes the differentiation of glutamatergic synapses and interacts with SLITRK3. Together, both proteins selectively regulate the development of inhibitory GABAergic synapses. Although no SNPs were identified as associated with OCD at genome-wide significance level, follow-up analyses of genome-wide association study (GWAS) signals from a previously published OCD study identified significant enrichment (P=0.0176). Secondary analyses of high-confidence interaction partners of DLGAP1 and GRIK2 (both showing evidence for association in our follow-up and the original GWAS study) revealed a trend of association (P=0.075) for a set of genes such as NEUROD6, SV2A, GRIA4, SLC1A2 and PTPRD. Analyses at the gene level revealed association of IQCK and C16orf88 (both P<1 × 10(-)(6), experiment-wide significant), as well as OFCC1 (P=6.29 × 10(-)(5)). The suggestive findings in this study await replication in larger samples.

Analysis of t(9;17)(q33.2;q25.3) chromosomal breakpoint regions and genetic association reveals novel candidate genes for bipolar disorder

OBJECTIVES

Breakpoints of chromosomal abnormalities facilitate identification of novel candidate genes for psychiatric disorders. Genome-wide significant evidence supports the linkage between chromosome 17q25.3 and bipolar disorder (BD). Co-segregation of translocation t(9;17)(q33.2;q25.3) with psychiatric disorders has been reported. We aimed to narrow down these chromosomal breakpoint regions and to investigate the associations between single nucleotide polymorphisms within these regions and BD as well as schizophrenia (SZ) in large genome-wide association study samples.

METHODS

We cross-linked Danish psychiatric and cytogenetic case registers to identify an individual with both t(9;17)(q33.2;q25.3) and BD. Fluorescent in situ hybridization was employed to map the chromosomal breakpoint regions of this proband. We accessed the Psychiatric Genomics Consortium BD (n = 16,731) and SZ (n = 21,856) data. Genetic associations between these disorders and single nucleotide polymorphisms within these breakpoint regions were analysed by BioQ, FORGE, and RegulomeDB programmes.

RESULTS

Four protein-coding genes [coding for (endonuclease V (ENDOV), neuronal pentraxin I (NPTX1), ring finger protein 213 (RNF213), and regulatory-associated protein of mammalian target of rapamycin (mTOR) (RPTOR)] were found to be located within the 17q25.3 breakpoint region. NPTX1 was significantly associated with BD (p = 0.004), while ENDOV was significantly associated with SZ (p = 0.0075) after Bonferroni correction.

CONCLUSIONS

Prior linkage evidence and our findings suggest NPTX1 as a novel candidate gene for BD.

Predicting the phenotypic values of physiological traits using SNP genotype and gene expression data in mice

Predicting phenotypes using genome-wide genetic variation and gene expression data is useful in several fields, such as human biology and medicine, as well as in crop and livestock breeding. However, for phenotype prediction using gene expression data for mammals, studies remain scarce, as the available data on gene expression profiling are currently limited. By integrating a few sources of relevant data that are available in mice, this study investigated the accuracy of phenotype prediction for several physiological traits. Gene expression data from two tissues as well as single nucleotide polymorphisms (SNPs) were used. For the studied traits, the variance of the effects of the expression levels was more likely to differ among the genes than were the effects of SNPs. For the glucose concentration, the total cholesterol amount, and the total tidal volume, the accuracy by cross validation tended to be higher when the gene expression data rather than the SNP genotype data were used, and a statistically significant increase in the accuracy was obtained when the gene expression data from the liver were used alone or jointly with the SNP genotype data. For these traits, there were no additional gains in accuracy from using the gene expression data of both the liver and lung compared to that of individual use. The accuracy of prediction using genes that were selected differently was examined; the use of genes with a higher tissue specificity tended to result in an accuracy that was similar to or greater than that associated with the use of all of the available genes for traits such as the glucose concentration and total cholesterol amount. Although relatively few animals were evaluated, the current results suggest that gene expression levels could be used as explanatory variables. However, further studies are essential to confirm our findings using additional animal samples.

Beyond GWAS in COPD: probing the landscape between gene-set associations, genome-wide associations and protein-protein interaction networks

OBJECTIVES

To use a systems biology approach to integrate genotype and protein-protein interaction (PPI) data to identify disease network modules associated with chronic obstructive pulmonary disease (COPD) and to perform traditional pathway analysis.

METHODS

We utilized a standard gene-set association approach (FORGE) using gene-based association analysis and gene-set definitions from the molecular signatures database (MSigDB). As a discovery step, we analyzed GWAS results from 2 well-characterized COPD cohorts: COPDGene and GenKOLS. We used a third well-characterized COPD case-control cohort for replication: ECLIPSE. Next, we used dmGWAS, a method that integrates GWAS results with PPI, to identify COPD disease modules.

RESULTS

No gene-sets reached experiment-wide significance in either discovery population. We identified a consensus network of 10 genes identified in modules by integrating GWAS results with PPI that replicated in COPDGene, GenKOLS, and ECLIPSE. Members of 4 gene-sets were enriched among these 10 genes: (i) lung adenocarcinoma tumor-sequencing genes, (ii) IL-7 pathway genes, (iii) kidney cell response to arsenic, and (iv) CD4 T-cell responses. Further, several genes have also been associated with pathophysiology relevant to COPD including KCNK3, NEDD4L, and RIN3. In particular, KCNK3 has been associated with pulmonary arterial hypertension, a common complication in advanced COPD.

CONCLUSION

We report a set of new genes that may influence the etiology of COPD that would not have been identified using traditional GWAS and pathway analyses alone.

Genome-wide pathway analysis of a genome-wide association study on Alzheimer's disease

The aims of this study were to identify candidate single nucleotide polymorphisms (SNPs) and mechanisms of Alzheimer's disease (AD) and to generate SNP to gene to pathway hypotheses. An AD genome-wide association study (GWAS) dataset that included 370,542 SNPs in 1,000 cases and 1,000 controls of European descent was used in this study. Identify Candidate Causal SNPs and Pathway (ICSNPathway) analysis was applied to the GWAS dataset. ICSNPathway analysis identified 3 candidate SNPs and 2 pathways, which provided 3 hypothetical biological mechanisms. The strongest hypothetical biological mechanism was rs8076604 [non-synonymous coding (deleterious)] to MYO18A to negative regulation of programmed cell death [nominal P < 0.001, false discovery rate (FDR) <0.043]. The second was rs2811226 (regulatory region) to ANXA1 to negative regulation of programmed cell death (nominal P < 0.001, FDR 0.043). The third was rs3734166 (non-synonymous coding) to CDC25C to M phase of the mitotic cell cycle (nominal P < 0.001, FDR 0.049). By applying the ICSNPathway analysis to the AD GWAS meta-analysis data, three candidate SNPs, three genes (MYO18A, ANXA1, CDC25C), 2 pathways involving negative regulation of programmed cell death and 1 pathway involving the M phase of the mitotic cell cycle were identified, which may contribute to AD susceptibility.

Integrated pathway-based approach identifies association between genomic regions at CTCF and CACNB2 and schizophrenia

In the present study, an integrated hierarchical approach was applied to: (1) identify pathways associated with susceptibility to schizophrenia; (2) detect genes that may be potentially affected in these pathways since they contain an associated polymorphism; and (3) annotate the functional consequences of such single-nucleotide polymorphisms (SNPs) in the affected genes or their regulatory regions. The Global Test was applied to detect schizophrenia-associated pathways using discovery and replication datasets comprising 5,040 and 5,082 individuals of European ancestry, respectively. Information concerning functional gene-sets was retrieved from the Kyoto Encyclopedia of Genes and Genomes, Gene Ontology, and the Molecular Signatures Database. Fourteen of the gene-sets or pathways identified in the discovery dataset were confirmed in the replication dataset. These include functional processes involved in transcriptional regulation and gene expression, synapse organization, cell adhesion, and apoptosis. For two genes, i.e. CTCF and CACNB2, evidence for association with schizophrenia was available (at the gene-level) in both the discovery study and published data from the Psychiatric Genomics Consortium schizophrenia study. Furthermore, these genes mapped to four of the 14 presently identified pathways. Several of the SNPs assigned to CTCF and CACNB2 have potential functional consequences, and a gene in close proximity to CACNB2, i.e. ARL5B, was identified as a potential gene of interest. Application of the present hierarchical approach thus allowed: (1) identification of novel biological gene-sets or pathways with potential involvement in the etiology of schizophrenia, as well as replication of these findings in an independent cohort; (2) detection of genes of interest for future follow-up studies; and (3) the highlighting of novel genes in previously reported candidate regions for schizophrenia.

Genome-wide pathway analysis of breast cancer

The aim of this study was to identify candidate single-nucleotide polymorphisms (SNPs) that might affect susceptibility to breast cancer and then elucidate their potential mechanisms and generate SNP-to-gene-to-pathway hypotheses. A genome-wide association study (GWAS) dataset of breast cancer that included 453,852 SNPs from 1,145 breast cancer patients and 1,142 control subjects of European descent was used in this study. The identify candidate causal SNPs and pathways (ICSNPathway) method was applied to the GWAS dataset. ICSNPathway analysis identified 16 candidate SNPs, 13 genes, and 7 pathways, which together revealed 7 hypothetical biological mechanisms. The strongest hypothetical biological mechanism was that rs3168891 and rs2899849 alter the role of MBIP in the inactivation of mitogen-activated protein kinase (MAPK) (p < 0.001; false discovery rate (FDR) = 0.038). The second strongest mechanism was that rs2229714 modulates RPS6KA1 to affect its role in growth hormone signaling (p = 0.001; FDR = 0.039). The third strongest mechanism was that rs2230394 modulates ITGB1 to regulate the PTEN pathway and hsa04360 (axon guidance pathway) (p < 0.001; FDR = 0.039, 0.041). Use of the ICSNPathway to analyze breast cancer GWAS data identified 16 candidate SNPs, 13 genes (including MBIP, RPS6KA1, and ITGB1), and 7 pathways that might contribute to the susceptibility of patients to breast cancer.

Genome-wide pathway analysis in neuroblastoma

The aim of this study was to identify candidate single-nucleotide polymorphisms (SNPs) that might play a role in susceptibility to neuroblastoma, elucidate their potential mechanisms, and generate SNP-to-gene-to-pathway hypotheses. A genome-wide association study (GWAS) dataset of neuroblastoma that included 442,976 SNPs from 1,627 neuroblastoma patients and 3,254 control subjects of European descent was used in this study. The identify candidate causal SNPs and pathways (ICSNPathway) analysis was applied to the GWAS dataset. ICSNPathway analysis identified 15 candidate SNPs, 10 genes, and 31 pathways, which revealed 10 hypothetical biological mechanisms. The strongest hypothetical biological mechanism was one wherein SNPrs40401 modulates the role of IL3 in several pathways and conditions, including the stem pathway, asthma (hsa05310), the dendritic cell pathway, and development (0.001 < p < 0.004; 0.001 < FDR < 0.033). The second strongest mechanism identified was that in which rs1048108 and rs16852600 alter the function of BARD1, which negatively regulates developmental process and modulates processes including cell development and programmed cell death (0.001 < p < 0.004; 0.001 < FDR < 0.033). The third mechanism identified was one wherein rs1939212 modulated CFL1, resulting in negative regulation of development, cell death, neural crest cell migration, and apoptosis (0.001 < p < 0.004; 0.001 < FDR < 0.033). By using the ICSNPathway to analyze neuroblastoma GWAS data, 15 candidate SNPs, 10 genes including IL3, BARD1, and CFL, and 31 pathways were identified that might contribute to the susceptibility of patients to neuroblastoma.

Pathway analysis of genome-wide association study on asthma

OBJECTIVE

The aims of this study were to identify the candidate causal single nucleotide polymorphisms (SNPs) and candidate causal mechanisms of asthma and to generate SNP to gene to pathway hypotheses.

METHODS

SNPs that met a threshold of p ≤ 0.001 in a genome-wide association study (GWAS) dataset of asthma, which included 292,443 SNPs in 473 asthma cases and 1892 controls, were used in the present study. Identify candidate causal SNPs and pathway (ICSNPathway) analysis was applied to this dataset.

RESULTS

ICSNPathway analysis identified four candidate causal SNPs, four genes, and 21 candidate causal pathways, which in total provided four hypothetical biologic mechanisms: (1) rs7192 (nonsynonymous coding) to HLA-DRA to 21 pathways, such as, the role of eosinophils in the chemokine network of allergy, Th1/Th2 differentiation, and asthma (nominal p ≤ 0.001, FDR p ≤ 0.01); (2) rs20541 (nonsynonymous coding) to IL13 to asthma and cytokines and inflammatory response (nominal p<0.001, FDR p ≤ 0.008); (3) rs1058808 (frameshift coding) to ERBB2 to transmembrane receptor activity (nominal p=0.001, FDR p=0.01); (4) rs17350764 (nonsynonymous coding (deleterious)) to OR52J3 to transmembrane receptor activity (nominal p=0.001, FDR p=0.01).

CONCLUSION

By applying ICSNPathway analysis to asthma GWAS data, we found four candidate causal SNPs, four genes involving HLA-DRA and IL-13, and four hypotheses, which may contribute to asthma susceptibility.

Genome-wide pathway analysis of a genome-wide association study on psoriasis and Behcet's disease

The aim of this study was to identify candidate causal single nucleotide polymorphisms (SNPs) and candidate causal mechanisms of psoriasis and Behcets's disease (BD) and to generate an SNP → gene → pathway hypothesis. A psoriasis genome-wide association study (GWAS) dataset that included 436,192 SNPs in 1,409 psoriasis cases and 1,436 controls of European descent and a BD GWAS dataset that contained 310,324 SNPs in 1,215 BD cases and 1,278 controls were used in this study. Identify candidate causal SNPs and pathways (ICSNPathway) analysis was applied to the GWAS datasets. ICSNPathway analysis identified 15 candidate causal SNPs and 28 candidate causal pathways. The top five candidate causal SNPs were rs1063478 (P = 1.45E-10), rs8084 (P = 2.20E-08), rs7192 (P = 5.18E-08), rs20541 (P = 5.30E-06), and rs1130838 (P = 5.65E-06), which with the exception of rs20541 [interleukin (IL)-13] are at human leukocyte antigen (HLA) loci. These candidate causal SNPs and pathways provided ten hypothetical biological mechanisms. The most strongly associated pathway concerned HLA. When HLA loci were excluded, ICSNPathway analysis provided one hypothetical biological mechanism. rs20541 (non_synonymous_coding) → IL-13 → dendritic cell involvement in the regulation of Th1 and Th2 development, and the GATA3 pathway. ICSNPathway analysis identified four candidate causal SNPs, eleven candidate causal pathways, and three hypothetical biological mechanisms. One of them was as follows: rs2072895 (non_synonymous_coding & splice-site) and rs2735059 (non_synonymous_coding) → HLA-F → type I diabetes mellitus, antigen processing and presentation, and autoimmune thyroid disease. The application of ICSNPathway analysis to GWAS dataset of psoriasis and BD resulted in the identification of candidate causal SNPs and candidate pathways that might contribute to psoriasis susceptibility.