scholarly journals The harmonic mean p-value for combining dependent tests

2019 ◽  
Vol 116 (4) ◽  
pp. 1195-1200 ◽  
Author(s):  
Daniel J. Wilson
Keyword(s):  
Multiple Testing ◽  
Statistical Power ◽  
Scientific Discovery ◽  
Association Studies ◽  
Harmonic Mean ◽  
P Value ◽  
Genome Wide Association Studies ◽  
Familywise Error Rate ◽  
Significance Threshold ◽  
Genome Wide

Analysis of “big data” frequently involves statistical comparison of millions of competing hypotheses to discover hidden processes underlying observed patterns of data, for example, in the search for genetic determinants of disease in genome-wide association studies (GWAS). Controlling the familywise error rate (FWER) is considered the strongest protection against false positives but makes it difficult to reach the multiple testing-corrected significance threshold. Here, I introduce the harmonic mean p-value (HMP), which controls the FWER while greatly improving statistical power by combining dependent tests using generalized central limit theorem. I show that the HMP effortlessly combines information to detect statistically significant signals among groups of individually nonsignificant hypotheses in examples of a human GWAS for neuroticism and a joint human–pathogen GWAS for hepatitis C viral load. The HMP simultaneously tests all ways to group hypotheses, allowing the smallest groups of hypotheses that retain significance to be sought. The power of the HMP to detect significant hypothesis groups is greater than the power of the Benjamini–Hochberg procedure to detect significant hypotheses, although the latter only controls the weaker false discovery rate (FDR). The HMP has broad implications for the analysis of large datasets, because it enhances the potential for scientific discovery.

scholarly journals The harmonic mean p-value for combining dependent tests

2017 ◽  
Author(s):  
Daniel J. Wilson
Keyword(s):  
Multiple Testing ◽  
Statistical Power ◽  
Scientific Discovery ◽  
Association Studies ◽  
Harmonic Mean ◽  
P Value ◽  
Genome Wide Association Studies ◽  
Genetic Determinants ◽  
Significance Threshold ◽  
Genome Wide

Analysis of ‘big data’ frequently involves statistical comparison of millions of competing hypotheses to discover hidden processes underlying observed patterns of data, for example in the search for genetic determinants of disease in genome-wide association studies (GWAS). Controlling the family-wise error rate (FWER) is considered the strongest protection against false positives, but makes it difficult to reach the multiple testing-corrected significance threshold. Here I introduce the harmonic mean p-value (HMP) which controls the FWER while greatly improving statistical power by combining dependent tests using generalized central limit theorem. I show that the HMP easily combines information to detect statistically significant signals among groups of individually nonsignificant hypotheses in examples of a human GWAS for neuroticism and a joint human-pathogen GWAS for hepatitis C viral load. The HMP simultaneously tests all combinations of hypotheses, allowing the smallest groups of hypotheses that retain significance to be sought. The power of the HMP to detect significant hypothesis groups is greater than the power of the Benjamini-Hochberg procedure to detect significant hypotheses, even though the latter only controls the weaker false discovery rate (FDR). The HMP has broad implications for the analysis of large datasets because it enhances the potential for scientific discovery.

scholarly journals An approach to gene-based testing accounting for dependence of tests among nearby genes

2021 ◽  
Author(s):  
Ronald J Yurko ◽  
Kathryn Roeder ◽  
Bernie Devlin ◽  
Max G'Sell
Keyword(s):  
Multiple Testing ◽  
Association Studies ◽  
Autism Spectrum ◽  
P Value ◽  
Genome Wide Association Studies ◽  
Strongly Correlated ◽  
Test Statistics ◽  
Test Statistic ◽  
Genome Wide ◽  
Insight Into

In genome-wide association studies (GWAS), it has become commonplace to test millions of SNPs for phenotypic association. Gene-based testing can improve power to detect weak signal by reducing multiple testing and pooling signal strength. While such tests account for linkage disequilibrium (LD) structure of SNP alleles within each gene, current approaches do not capture LD of SNPs falling in different nearby genes, which can induce correlation of gene-based test statistics. We introduce an algorithm to account for this correlation. When a gene's test statistic is independent of others, it is assessed separately; when test statistics for nearby genes are strongly correlated, their SNPs are agglomerated and tested as a locus. To provide insight into SNPs and genes driving association within loci, we develop an interactive visualization tool to explore localized signal. We demonstrate our approach in the context of weakly powered GWAS for autism spectrum disorder, which is contrasted to more highly powered GWAS for schizophrenia and educational attainment. To increase power for these analyses, especially those for autism, we use adaptive p-value thresholding (AdaPT), guided by high-dimensional metadata modeled with gradient boosted trees, highlighting when and how it can be most useful. Notably our workflow is based on summary statistics.

scholarly journals Optimal Genomic Control in Large-scale Genetic Associations for Binary Diseases

2021 ◽  
Author(s):  
Runqing Yang ◽  
Yuxin Song ◽  
Li Jiang ◽  
Zhiyu Hao ◽  
Runqing Yang
Keyword(s):  
Multiple Testing ◽  
Statistical Power ◽  
Large Scale ◽  
Association Studies ◽  
Joint Analysis ◽  
Genome Wide Association Studies ◽  
Genetic Associations ◽  
Genomic Heritability ◽  
Large Scale Data ◽  
Genome Wide

Abstract Complex computation and approximate solution hinder the application of generalized linear mixed models (GLMM) into genome-wide association studies. We extended GRAMMAR to handle binary diseases by considering genomic breeding values (GBVs) estimated in advance as a known predictor in genomic logit regression, and then controlled polygenic effects by regulating downward genomic heritability. Using simulations and case analyses, we showed in optimizing GRAMMAR, polygenic effects and genomic controls could be evaluated using the fewer sampling markers, which extremely simplified GLMM-based association analysis in large-scale data. In addition, joint analysis for quantitative trait nucleotide (QTN) candidates chosen by multiple testing offered significant improved statistical power to detect QTNs over existing methods.

scholarly journals The paltry power of priors versus populations

2019 ◽  
Author(s):  
Jianan Zhan ◽  
Dan E. Arking ◽  
Joel S. Bader
Keyword(s):  
Population Size ◽  
Multiple Testing ◽  
Association Studies ◽  
Genome Wide Association ◽  
P Value ◽  
Genome Wide Association Studies ◽  
Significant Finding ◽  
Rna Seq ◽  
Test Power ◽  
Genome Wide

AbstractBiological experiments often involve hypothesis testing at the scale of thousands to millions of tests. Alleviating the multiple testing burden has been a goal of many methods designed to boost test power by focusing tests on the alternative hypotheses most likely to be true. Very often, these methods either explicitly or implicitly make use of prior probabilities that bias significance for favored sets thought to be enriched for significant finding. Nevertheless, most genomics experiments, and in particular genome-wide association studies (GWAS), still use traditional univariate tests rather than more sophisticated approaches. Here we use GWAS to demonstrate why unbiased tests remain in favor. We calculate test power assuming perfect knowledge of a prior distribution and then derive the population size increase required to provided the same boost without a prior. We show that population size is exponentially more important than prior, providing a rigorous explanation for the observed avoidance of prior-based methods.Author summaryBiological experiments often test thousands to millions of hypotheses. Gene-based tests for human RNA-Seq data, for example, involve approximately 20,000; genome-wide association studies (GWAS) involve about 1 million effective tests. The conventional approach is to perform individual tests and then apply a Bonferroni correction to account for multiple testing. This approach implies a single-test p-value of 2.5 × 10−6 for RNA-Seq experiments, and a p-value of 5 × 10−8 for GWAS, to control the false-positive rate at a conventional value of 0.05. Many methods have been proposed to alleviate the multiple-testing burden by incorporating a prior probability that boosts the significance for a subset of candidate genes or variants. At the extreme limit, only the candidate set is tested, corresponding to a decreased multiple testing burden. Despite decades of methods development, prior-based tests have not been generally used. Here we compare the power increase possible with a prior with the increase possible with a much simpler strategy of increasing a study size. We show that increasing the population size is exponentially more valuable than increasing the strength of prior, even when the true prior is known exactly. These results provide a rigorous explanation for the continued use of simple, robust methods rather than more sophisticated approaches.

scholarly journals FORGE: multivariate calculation of gene-wide p-values from Genome-Wide Association Studies Authors and Affiliations

2015 ◽  
Author(s):  
Inti Inal Pedroso ◽  
Michael R Barnes ◽  
Anbarasu Lourdusamy ◽  
Ammar Al-Chalabi ◽  
Gerome Breen
Keyword(s):  
Statistical Power ◽  
Association Studies ◽  
Single Point ◽  
Genome Wide Association ◽  
P Value ◽  
Disease Genes ◽  
Snp Analysis ◽  
Genome Wide Association Studies ◽  
P Values ◽  
Genome Wide

Genome-wide association studies (GWAS) have proven a valuable tool to explore the genetic basis of many traits. However, many GWAS lack statistical power and the commonly used single-point analysis method needs to be complemented to enhance power and interpretation. Multivariate region or gene-wide association are an alternative, allowing for identification of disease genes in a manner more robust to allelic heterogeneity. Gene-based association also facilitates systems biology analyses by generating a single p-value per gene. We have designed and implemented FORGE, a software suite which implements a range of methods for the combination of p-values for the individual genetic variants within a gene or genomic region. The software can be used with summary statistics (marker ids and p-values) and accepts as input the result file formats of commonly used genetic association software. When applied to a study of Crohn's disease susceptibility, it identified all genes found by single SNP analysis and additional genes identified by large independent meta-analysis. FORGE p-values on gene-set analyses highlighted association with the Jak-STAT and cytokine signalling pathways, both previously associated with CD. We highlight the software's main features, its future development directions and provide a comparison with alternative available software tools. FORGE can be freely accessed at https://github.com/inti/FORGE.

scholarly journals Polymorphic variants in Sweet and Umami taste receptor genes and birthweight

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Riccardo Farinella ◽  
Ilaria Erbi ◽  
Alice Bedini ◽  
Sara Donato ◽  
Manuel Gentiluomo ◽  
...  
Keyword(s):  
Birth Weight ◽  
Multiple Testing ◽  
Association Studies ◽  
Taste Receptor ◽  
Adult Life ◽  
P Value ◽  
Genome Wide Association Studies ◽  
Long Term Effects ◽  
Umami Taste ◽  
Genome Wide

AbstractThe first thousand days of life from conception have a significant impact on the health status with short, and long-term effects. Among several anthropometric and maternal lifestyle parameters birth weight plays a crucial role on the growth and neurological development of infants. Recent genome wide association studies (GWAS) have demonstrated a robust foetal and maternal genetic background of birth weight, however only a small proportion of the genetic hereditability has been already identified. Considering the extensive number of phenotypes on which they are involved, we focused on identifying the possible effect of genetic variants belonging to taste receptor genes and birthweight. In the human genome there are two taste receptors family the bitter receptors (TAS2Rs) and the sweet and umami receptors (TAS1Rs). In particular sweet perception is due to a heterodimeric receptor encoded by the TAS1R2 and the TAS1R3 gene, while the umami taste receptor is encoded by the TAS1R1 and the TAS1R3 genes. We observed that carriers of the T allele of the TAS1R1-rs4908932 SNPs showed an increase in birthweight compared to GG homozygotes Coeff: 87.40 (35.13–139.68) p-value = 0.001. The association remained significant after correction for multiple testing. TAS1R1-rs4908932 is a potentially functional SNP and is in linkage disequilibrium with another polymorphism that has been associated with BMI in adults showing the importance of this variant from the early stages of conception through all the adult life.

scholarly journals Covariate Adaptive Family-wise Error Rate Control for Genome-Wide Association Studies

Biometrika ◽  
2020 ◽  
Author(s):  
Huijuan Zhou ◽  
Xianyang Zhang ◽  
Jun Chen
Keyword(s):  
Error Rate ◽  
Rate Control ◽  
Multiple Testing ◽  
Statistical Power ◽  
Association Studies ◽  
Genome Wide Association ◽  
Genome Wide Association Studies ◽  
Nucleotide Polymorphisms ◽  
Family Wise Error Rate ◽  
Genome Wide

Abstract The family-wise error rate (FWER) has been widely used in genome-wide association studies. With the increasing availability of functional genomics data, it is possible to increase the detection power by leveraging these genomic functional annotations. Previous efforts to accommodate covariates in multiple testing focus on the false discovery rate control while covariate-adaptive FWER-controlling procedures remain under-developed. Here we propose a novel covariate-adaptive FWER-controlling procedure that incorporates external covariates which are potentially informative of either the statistical power or the prior null probability. An efficient algorithm is developed to implement the proposed method. We prove its asymptotic validity and obtain the rate of convergence through a perturbation-type argument. Our numerical studies show that the new procedure is more powerful than competing methods and maintains robustness across different settings. We apply the proposed approach to the UK Biobank data and analyze 27 traits with 9 million single-nucleotide polymorphisms tested for associations. Seventy-five genomic annotations are used as covariates. Our approach detects more genome-wide significant loci than other methods in 21 out of the 27 traits.

scholarly journals Optimal Genomic Control in Large-scale Genetic Associations for Binary Diseases

2021 ◽  
Author(s):  
Runqing Yang ◽  
Yuxin Song ◽  
Li Jiang ◽  
Zhiyu Hao ◽  
Runqing Yang
Keyword(s):  
Multiple Testing ◽  
Statistical Power ◽  
Large Scale ◽  
Association Studies ◽  
Joint Analysis ◽  
Genome Wide Association Studies ◽  
Genetic Associations ◽  
Genomic Heritability ◽  
Large Scale Data ◽  
Genome Wide

Abstract Complex computation and approximate solution hinder the application of generalized linear mixed models (GLMM) into genome-wide association studies. We extended GRAMMAR to handle binary diseases by considering genomic breeding values (GBVs) estimated in advance as a known predictor in genomic logit regression, and then controlled polygenic effects by regulating downward genomic heritability. Using simulations and case analyses, we showed in optimizing GRAMMAR, polygenic effects and genomic controls could be evaluated using the fewer sampling markers, which extremely simplified GLMM-based association analysis in large-scale data. In addition, joint analysis for quantitative trait nucleotide (QTN) candidates chosen by multiple testing offered significant improved statistical power to detect QTNs over existing methods.

scholarly journals Recently evolved human-specific methylated regions are enriched in schizophrenia signals

2017 ◽  
Author(s):  
Niladri Banerjee ◽  
Tatiana Polushina ◽  
Francesco Bettella ◽  
Sudheer Giddaluru ◽  
Vidar M. Steen ◽  
...  
Keyword(s):  
Human Evolution ◽  
Multiple Testing ◽  
Environmental Changes ◽  
Association Studies ◽  
Epigenetic Mark ◽  
Genome Wide Association Studies ◽  
Significance Threshold ◽  
Genome Wide ◽  
Genomic Regions ◽  
Human Specific

AbstractBackgroundOne explanation for the persistence of schizophrenia despite the reduced fertility of patients is that it is a by-product of recent human evolution. This hypothesis is supported by evidence suggesting that recently-evolved genomic regions in humans are involved in the genetic risk for schizophrenia. Using summary statistics from genome-wide association studies (GWAS) of schizophrenia and 11 other phenotypes, we tested for enrichment of association with GWAS traits in regions that have undergone methylation changes in the human lineage compared to Neanderthals and Denisovans, i.e. human-specific differentially methylated regions (DMRs). We used analytical tools that evaluate polygenic enrichment of a subset of genomic variants against all variants.ResultsSchizophrenia was the only trait in which DMR SNPs showed clear enrichment of association that passed the genome-wide significance threshold. The enrichment was not observed for Neanderthal or Denisovan DMRs. The enrichment seen in human DMRs is comparable to that for genomic regions tagged by Neanderthal Selective Sweep markers, and stronger than that for Human Accelerated Regions. The enrichment survives multiple testing performed through permutation (n=10,000) and bootstrapping (n=5,000) in INRICH (p<0.01). Some enrichment of association with height was observed at the gene level.ConclusionsRegions where DNA methylation modifications have changed during recent human evolution show enrichment of association with schizophrenia and possibly with height. Our study further supports the hypothesis that genetic variants conferring risk of schizophrenia co-occur in genomic regions that have changed as the human species evolved. Since methylation is an epigenetic mark, potentially mediated by environmental changes, our results also suggest that interaction with the environment might have contributed to that association.

scholarly journals MetaPhat: Detecting and decomposing multivariate associations from univariate genome-wide association statistics

2019 ◽  
Author(s):  
Jake Lin ◽  
Rubina Tabassum ◽  
Samuli Ripatti ◽  
Matti Pirinen
Keyword(s):  
Statistical Power ◽  
Association Studies ◽  
Information Criterion ◽  
Genome Wide Association ◽  
P Value ◽  
Genome Wide Association Studies ◽  
R Software ◽  
Testing Tools ◽  
Genome Wide ◽  
Lipid Species

AbstractBackgroundMultivariate testing tools that integrate multiple genome-wide association studies (GWAS) have become important as the number of phenotypes gathered from study cohorts and biobanks has increased. While these tools have been shown to boost statistical power considerably over univariate tests, an important remaining challenge is to interpret which traits are driving the multivariate association and which traits are just passengers with minor contributions to the genotype-phenotypes association statistic.ResultsWe introduce MetaPhat, a novel bioinformatics tool to conduct GWAS of multivariate and correlated traits using univariate summary results and to decompose multivariate associations into sets of central traits based on intuitive trace plots that visualize Bayesian Information Criterion (BIC) and P-value statistics of multivariate association models. We validate MetaPhat with Global Lipids Genetics Consortium GWAS results, and we apply MetaPhat to univariate GWAS results for 21 heritable and correlated polyunsaturated lipid species from 2,045 Finnish samples, detecting 7 independent loci associated with a cluster of lipid species. In most cases, we are able to decompose these multivariate associations to only 3-5 central traits out of all 21 traits included in the analyses. We release MetaPhat as an open source tool written in Python with built-in support for multi-processing, quality control, clumping and intuitive visualizations using the R software.ConclusionsMetaPhat efficiently decomposes associations between multivariate phenotypes and genetic variants into smaller sets of central traits and improves the interpretation and specificity of genome-phenome associations.MetaPhat is available under the MIT license at: https://sourceforge.net/projects/meta-pheno-association-tracer

Sign in / Sign up

Export Citation Format

Share Document