A comparison of aggregate p-value methods and multivariate statistics for self-contained tests of metabolic pathway analysis.
Ontology highlight
ABSTRACT:
SUBMITTER: Mitchell MW
PROVIDER: S-EPMC4415974 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Complex diseases are usually associated with multiple correlated phenotypes, and the analysis of composite scores or disease status may not fully capture the complexity (or multidimensionality). Joint analysis of multiple disease-related phenotypes in genetic tests could potentially increase power to detect association of a disease with common SNPs (or genes). Gene-based tests are designed to identify genes containing multiple risk variants that individually are weakly associated with a univariate trait. We combined three multivariate association tests (O'Brien method, TATES, and MultiPhen) with two gene-based association tests (GATES and VEGAS) and compared performance (type I error and power) of six multivariate gene-based methods using simulated data. Data (n = 2000) for genetic sequence and correlated phenotypes were simulated by varying causal variant proportions and phenotype correlations for various scenarios. These simulations showed that two multivariate association tests (TATES and MultiPhen, but not O'Brien) paired with VEGAS have inflated type I error in all scenarios, while the three multivariate association tests paired with GATES have correct type I error. MultiPhen paired with GATES has higher power than competing methods if the correlations among phenotypes are low (r < 0.57). We applied these gene-based association methods to a GWAS dataset from the Alzheimer's Disease Genetics Consortium containing three neuropathological traits related to Alzheimer disease (neuritic plaque, neurofibrillary tangles, and cerebral amyloid angiopathy) measured in 3500 autopsied brains. Gene-level significant evidence (P < 2.7 × 10-6) was identified in a region containing three contiguous genes (TRAPPC12, TRAPPC12-AS1, ADI1) using O'Brien and VEGAS. Gene-wide significant associations were not observed in univariate gene-based tests.
| S-EPMC6461986 | biostudies-literature
Project description:Statistical inference in neuroimaging research often involves testing the significance of regression coefficients in a general linear model. In many applications, the researcher assumes a model of the form Y=?+X?+Z?+?, where Y is the observed brain signal, and X and Z contain explanatory variables that are thought to be related to the brain signal. The goal is to test the null hypothesis H0:?=0 with the nuisance parameters ? included in the model. Several nonparametric (permutation) methods have been proposed for this problem, and each method uses some variant of the F ratio as the test statistic. However, recent research suggests that the F ratio can produce invalid permutation tests of H0:?=0 when the ? terms are heteroscedastic (i.e., have non-constant variance), which can occur for a variety of reasons. This study compares the classic F test statistic to the robust W (Wald) test statistic using eight different permutation methods. The results reveal that permutation tests using the F ratio can produce accurate results when the errors are homoscedastic, but high false positive rates when the errors are heteroscedastic. In contrast, permutation tests using the W test statistic produced valid results when the errors were homoscedastic, and asymptotically valid results when the errors were heteroscedastic. In the situation with homoscedastic errors, permutation tests using the W statistic showed slightly reduced power compared to the F statistic, but the difference disappeared as the sample size n increased. Consequently, the W test statistic is recommended for robust nonparametric hypothesis tests of regression coefficients in neuroimaging research.
| S-EPMC6765412 | biostudies-literature
Project description:The analysis of differentially expressed gene sets became a routine in the analyses of gene expression data. There is a multitude of tests available, ranging from aggregation tests that summarize gene-level statistics for a gene set to true multivariate tests, accounting for intergene correlations. Most of them detect complex departures from the null hypothesis but when the null hypothesis is rejected, the specific alternative leading to the rejection is not easily identifiable.In this article we compare the power and Type I error rates of minimum-spanning tree (MST)-based non-parametric multivariate tests with several multivariate and aggregation tests, which are frequently used for pathway analyses. In our simulation study, we demonstrate that MST-based tests have power that is for many settings comparable with the power of conventional approaches, but outperform them in specific regions of the parameter space corresponding to biologically relevant configurations. Further, we find for simulated and for gene expression data that MST-based tests discriminate well against shift and scale alternatives. As a general result, we suggest a two-step practical analysis strategy that may increase the interpretability of experimental data: first, apply the most powerful multivariate test to find the subset of pathways for which the null hypothesis is rejected and second, apply MST-based tests to these pathways to select those that support specific alternative hypotheses.gvglazko@uams.edu or yrahmatallah@uams.eduSupplementary data are available at Bioinformatics online.
| S-EPMC3509490 | biostudies-literature
Project description:Joint association analysis of multiple traits in a genome-wide association study (GWAS), i.e. a multivariate GWAS, offers several advantages over analyzing each trait in a separate GWAS. In this study we directly compared a number of multivariate GWAS methods using simulated data. We focused on six methods that are implemented in the software packages PLINK, SNPTEST, MultiPhen, BIMBAM, PCHAT and TATES, and also compared them to standard univariate GWAS, analysis of the first principal component of the traits, and meta-analysis of univariate results. We simulated data (N = 1000) for three quantitative traits and one bi-allelic quantitative trait locus (QTL), and varied the number of traits associated with the QTL (explained variance 0.1%), minor allele frequency of the QTL, residual correlation between the traits, and the sign of the correlation induced by the QTL relative to the residual correlation. We compared the power of the methods using empirically fixed significance thresholds (α = 0.05). Our results showed that the multivariate methods implemented in PLINK, SNPTEST, MultiPhen and BIMBAM performed best for the majority of the tested scenarios, with a notable increase in power for scenarios with an opposite sign of genetic and residual correlation. All multivariate analyses resulted in a higher power than univariate analyses, even when only one of the traits was associated with the QTL. Hence, use of multivariate GWAS methods can be recommended, even when genetic correlations between traits are weak.
| S-EPMC3999149 | biostudies-literature
Project description:OBJECTIVE: Genome-wide association studies have successfully elucidated the genetic background of complex diseases, but X chromosomal data have usually not been analyzed. A reason for this is that there is no consensus approach for the analysis taking into account the specific features of X chromosomal data. This contribution evaluates test statistics proposed for X chromosomal markers regarding type I error frequencies and power. METHODS: We performed extensive simulation studies covering a wide range of different settings. Besides characteristics of the general population, we investigated sex-balanced or unbalanced sampling procedures as well as sex-specific effect sizes, allele frequencies and prevalence. Finally, we applied the test statistics to an association data set on Crohn's disease. RESULTS: Simulation results imply that in addition to standard quality control, sex-specific allele frequencies should be checked to control for type I errors. Furthermore, we observed distinct differences in power between test statistics which are determined by sampling design and sex specificity of effect sizes. Analysis of the Crohn's disease data detects two previously unknown genetic regions on the X chromosome. CONCLUSION: Although no test is uniformly most powerful under all settings, recommendations are offered as to which test performs best under certain conditions.
| S-EPMC3089425 | biostudies-literature
Project description:Computationally efficient statistical tests are needed in association testing of large scale genetic markers for survival outcomes. In this study, we explore several test statistics based on the Cox proportional hazards model for survival data. First, we consider the classical partial likelihood-based Wald and score tests. A revised way to compute the score statistics is explored to improve the computational efficiency. Next, we propose a Cox-Snell residual-based score test, which allows us to handle the controlling variables more conveniently. We also illustrated the incorporation of these three tests into a permutation procedure to adjust for the multiple testing. In addition, we examine a simulation-based approach proposed by Lin (2005) to adjust for multiple testing. We presented the comparison of these four statistics in terms of type I error, power, family-wise error rate, and computational efficiency under various scenarios via extensive simulation.
| S-EPMC3985281 | biostudies-literature
Project description:It has been suggested that pathway analysis can complement single-SNP analysis in exploring genomewide association data. Pathway analysis incorporates the available biological knowledge of genes and SNPs and is expected to improve the chances of revealing the underlying genetic architecture of complex traits. Methods for pathway analysis can be classified as competitive (enrichment) or self-contained (association) according to the hypothesis tested. Although association tests are statistically more powerful than enrichment tests they can be difficult to calibrate because biases in analysis accumulate across multiple SNPs or genes. Furthermore, enrichment tests can be more scientifically relevant than association tests, as they detect pathways with relatively more evidence for association than the remaining genes. Here we show how some well known association tests can be simply adapted to test for enrichment, and compare their performance to some established enrichment tests. We propose versions of the Adaptive Rank Truncated Product (ARTP), Tail Strength Measure and Fisher's combination of p-values for testing the enrichment null hypothesis. We compare the behaviour of these proposed methods with the established Hypergeometric Test and Gene-Set Enrichment Analysis (GSEA). The results of the simulation study show that the modified version of the ARTP method has generally the best performance across the situations considered. The methods were also applied for finding enriched pathways for body mass index (BMI) and platelet function phenotypes. The pathway analysis of BMI identified the Vasoactive Intestinal Peptide pathway as significantly associated with BMI. This pathway has been previously reported as associated with BMI and the risk of obesity. The ARTP method was the method that identified the largest number of enriched pathways across all tested pathway databases and phenotypes. The simulation and data application results are in agreement with previous work on association tests and suggests that the ARTP should be preferred for both enrichment and association testing.
| S-EPMC3409204 | biostudies-literature
Project description:Gene set methods aim to assess the overall evidence of association of a set of genes with a phenotype, such as disease or a quantitative trait. Multiple approaches for gene set analysis of expression data have been proposed. They can be divided into two types: competitive and self-contained. Benefits of self-contained methods include that they can be used for genome-wide, candidate gene, or pathway studies, and have been reported to be more powerful than competitive methods. We therefore investigated ten self-contained methods that can be used for continuous, discrete and time-to-event phenotypes. To assess the power and type I error rate for the various previously proposed and novel approaches, an extensive simulation study was completed in which the scenarios varied according to: number of genes in a gene set, number of genes associated with the phenotype, effect sizes, correlation between expression of genes within a gene set, and the sample size. In addition to the simulated data, the various methods were applied to a pharmacogenomic study of the drug gemcitabine. Simulation results demonstrated that overall Fisher's method and the global model with random effects have the highest power for a wide range of scenarios, while the analysis based on the first principal component and Kolmogorov-Smirnov test tended to have lowest power. The methods investigated here are likely to play an important role in identifying pathways that contribute to complex traits.
| S-EPMC2941449 | biostudies-literature
Project description:The diet plays a major role in shaping gut microbiome composition and function in both humans and animals, and dietary intervention trials are often used to investigate and understand these effects. A plethora of statistical methods for analysing the differential abundance of microbial taxa exists, and new methods are constantly being developed, but there is a lack of benchmarking studies and clear consensus on the best multivariate statistical practices. This makes it hard for a biologist to decide which method to use. We compared the outcomes of generic multivariate ANOVA (ASCA and FFMANOVA) against statistical methods commonly used for community analyses (PERMANOVA and SIMPER) and methods designed for analysis of count data from high-throughput sequencing experiments (ALDEx2, ANCOM and DESeq2). The comparison is based on both simulated data and five published dietary intervention trials representing different subjects and study designs. We found that the methods testing differences at the community level were in agreement regarding both effect size and statistical significance. However, the methods that provided ranking and identification of differentially abundant operational taxonomic units (OTUs) gave incongruent results, implying that the choice of method is likely to influence the biological interpretations. The generic multivariate ANOVA tools have the flexibility needed for analysing multifactorial experiments and provide outputs at both the community and OTU levels; good performance in the simulation studies suggests that these statistical tools are also suitable for microbiome data sets.
| S-EPMC8601541 | biostudies-literature
Project description:The polygenic risk score (PRS) is calculated as the weighted sum of an individual's genotypes and their estimated effect sizes, which is often used to estimate an individual's genetic susceptibility to complex traits and disorders. It is well known that some complex human traits or disorders have sex differences in trait distributions, disease onset, progression, and treatment response, although the underlying mechanisms causing these sex differences remain largely unknown. PRSs for these traits are often based on Genome-Wide Association Studies (GWAS) data with both male and female samples included, ignoring sex differences. In this study, we present a benchmark study using both simulations with various combinations of genetic correlation and sample size ratios between sexes and real data to investigate whether combining sex-specific PRSs can outperform sex-agnostic PRSs on traits showing sex differences. We consider two types of PRS models in our study: single-population PRS models (PRScs, LDpred2) and multiple-population PRS models (PRScsx). For each trait or disorder, the candidate PRSs were calculated based on sex-specific GWAS data and sex-agnostic GWAS data. The simulation results show that applying LDpred2 or PRScsx to sex-specific GWAS data and then combining sex-specific PRSs leads to the highest prediction accuracy when the genetic correlation between sexes is low and the sample sizes for both sexes are balanced and large. Otherwise, the PRS generated by applying LDpred2 or PRScs to sex-agnostic GWAS data is more appropriate. If the sample sizes between sexes are not too small and very unbalanced, combining LDpred2-based sex-specific PRSs to predict on the sex with a larger sample size and combining PRScsx-based sex-specific PRSs to predict on the sex with a smaller size are the preferred strategies. For real data, we considered 19 traits from Genetic Investigation of ANthropometric Traits (GIANT) consortium studies and UK Biobank with both sex-specific GWAS data and sex-agnostic GWAS data. We found that for waist-to-hip ratio (WHR) related traits, accounting for sex differences and incorporating information from the opposite sex could help improve PRS prediction accuracy. Taken together, our findings in this study provide guidance on how to calculate the best PRS for sex-differentiated traits or disorders, especially as the sample size of GWASs grows in the future.
| S-EPMC9304553 | biostudies-literature