1
|
Chen Z. Robust tests for combining p-values under arbitrary dependency structures. Sci Rep 2022; 12:3158. [PMID: 35210502 PMCID: PMC8873210 DOI: 10.1038/s41598-022-07094-7] [Citation(s) in RCA: 6] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/07/2021] [Accepted: 02/11/2022] [Indexed: 12/16/2022] Open
Abstract
Recently Liu and Xie proposed a p-value combination test based on the Cauchy distribution (CCT). They showed that when the significance levels are small, CCT can control type I error rate and the resulting p-value can be simply approximated using a Cauchy distribution. One very special and attractive property of CCT is that it is applicable to situations where the p-values to be combined are dependent. However, in this paper, we show that under some conditions the commonly used MinP test is much more powerful than CCT. In addition, under some other situations, CCT is powerless at all. Therefore, we should use CCT with caution. We also proposed new robust p-value combination tests using a second MinP/CCT to combine the dependent p-values obtained from CCT and MinP applied to the original p-values. We call the new tests MinP-CCT-MinP (MCM) and CCT-MinP-CCT (CMC). We study the performance of the new tests by comparing them with CCT and MinP using comprehensive simulation study. Our study shows that the proposed tests, MCM and CMC, are robust and powerful under many conditions, and can be considered as alternatives of CCT or MinP.
Collapse
Affiliation(s)
- Zhongxue Chen
- Department of Epidemiology and Biostatistics, School of Public Health, Indiana University Bloomington, 1025 E. 7th Street, Bloomington, IN, 47405, USA.
| |
Collapse
|
2
|
Moore CM, Jacobson SA, Fingerlin TE. Power and Sample Size Calculations for Genetic Association Studies in the Presence of Genetic Model Misspecification. Hum Hered 2020; 84:256-271. [PMID: 32721961 DOI: 10.1159/000508558] [Citation(s) in RCA: 33] [Impact Index Per Article: 8.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/09/2020] [Accepted: 05/11/2020] [Indexed: 01/02/2023] Open
Abstract
INTRODUCTION When analyzing data from large-scale genetic association studies, such as targeted or genome-wide resequencing studies, it is common to assume a single genetic model, such as dominant or additive, for all tests of association between a given genetic variant and the phenotype. However, for many variants, the chosen model will result in poor model fit and may lack statistical power due to model misspecification. OBJECTIVE We develop power and sample size calculations for tests of gene and gene × environment interaction, allowing for misspecification of the true mode of genetic susceptibility. METHODS The power calculations are based on a likelihood ratio test framework and are implemented in an open-source R package ("genpwr"). RESULTS We use these methods to develop an analysis plan for a resequencing study in idiopathic pulmonary fibrosis and show that using a 2-degree of freedom test can increase power to detect recessive genetic effects while maintaining power to detect dominant and additive effects. CONCLUSIONS Understanding the impact of model misspecification can aid in study design and developing analysis plans that maximize power to detect a range of true underlying genetic effects. In particular, these calculations help identify when a multiple degree of freedom test or other robust test of association may be advantageous.
Collapse
Affiliation(s)
- Camille M Moore
- Center for Genes, Environment, and Health, National Jewish Health, Denver, Colorado, USA,
| | - Sean A Jacobson
- Center for Genes, Environment, and Health, National Jewish Health, Denver, Colorado, USA
| | - Tasha E Fingerlin
- Center for Genes, Environment, and Health, National Jewish Health, Denver, Colorado, USA
| |
Collapse
|
3
|
Zyla J, Kabacik S, O'Brien G, Wakil S, Al-Harbi N, Kaprio J, Badie C, Polanska J, Alsbeih G. Combining CDKN1A gene expression and genome-wide SNPs in a twin cohort to gain insight into the heritability of individual radiosensitivity. Funct Integr Genomics 2019; 19:575-585. [PMID: 30706161 PMCID: PMC6570669 DOI: 10.1007/s10142-019-00658-3] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.4] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/02/2018] [Revised: 12/12/2018] [Accepted: 01/09/2019] [Indexed: 12/15/2022]
Abstract
Individual variability in response to radiation exposure is recognised and has often been reported as important in treatment planning. Despite many efforts to identify biomarkers allowing the identification of radiation sensitive patients, it is not yet possible to distinguish them with certainty before the beginning of the radiotherapy treatment. A comprehensive analysis of genome-wide single-nucleotide polymorphisms (SNPs) and a transcriptional response to ionising radiation exposure in twins have the potential to identify such an individual. In the present work, we investigated SNP profile and CDKN1A gene expression in blood T lymphocytes from 130 healthy Caucasians with a complex level of individual kinship (unrelated, mono- or dizygotic twins). It was found that genetic variation accounts for 66% (95% CI 37-82%) of CDKN1A transcriptional response to radiation exposure. We developed a novel integrative multi-kinship strategy allowing investigating the role of genome-wide polymorphisms in transcriptomic radiation response, and it revealed that rs205543 (ETV6 gene), rs2287505 and rs1263612 (KLF7 gene) are significantly associated with CDKN1A expression level. The functional analysis revealed that rs6974232 (RPA3 gene), involved in mismatch repair (p value = 9.68e-04) as well as in RNA repair (p value = 1.4e-03) might have an important role in that process. Two missense polymorphisms with possible deleterious effect in humans were identified: rs1133833 (AKIP1 gene) and rs17362588 (CCDC141 gene). In summary, the data presented here support the validity of this novel integrative data analysis strategy to provide insights into the identification of SNPs potentially influencing radiation sensitivity. Further investigations in radiation response research at the genomic level should be therefore continued to confirm these findings.
Collapse
Affiliation(s)
- Joanna Zyla
- Data Mining Division, Faculty of Automatic Control, Electronic and Computer Science, Silesian University of Technology, Akademicka 16, 44-100, Gliwice, Poland
| | - Sylwia Kabacik
- Cellular Biology Group, Radiation Effects Department, Centre for Radiation, Chemical and Environmental Hazards, Public Health England, Chilton, Didcot, OX11 0RQ, UK
| | - Grainne O'Brien
- Cancer Mechanisms and Biomarkers Group, Radiation Effects Department, Centre for Radiation, Chemical and Environmental Hazards, Public Health England, Chilton, Didcot, OX11 0RQ, UK
| | - Salma Wakil
- Department of Genetics, King Faisal Specialist Hospital and Research Centre, Riyadh, 11211, Kingdom of Saudi Arabia
| | - Najla Al-Harbi
- Radiation Biology Section, Biomedical Physics Department, King Faisal Specialist Hospital and Research Centre, Riyadh, 11211, Kingdom of Saudi Arabia
| | - Jaakko Kaprio
- Department of Public Health and Institute for Molecular Medicine FIMM, University of Helsinki, 00140, Helsinki, Finland
| | - Christophe Badie
- Cancer Mechanisms and Biomarkers Group, Radiation Effects Department, Centre for Radiation, Chemical and Environmental Hazards, Public Health England, Chilton, Didcot, OX11 0RQ, UK
| | - Joanna Polanska
- Data Mining Division, Faculty of Automatic Control, Electronic and Computer Science, Silesian University of Technology, Akademicka 16, 44-100, Gliwice, Poland.
| | - Ghazi Alsbeih
- Radiation Biology Section, Biomedical Physics Department, King Faisal Specialist Hospital and Research Centre, Riyadh, 11211, Kingdom of Saudi Arabia
| |
Collapse
|
4
|
Chen Z, Liu Q, Wang K. A genetic association test through combining two independent tests. Genomics 2018; 111:1152-1159. [PMID: 30009923 DOI: 10.1016/j.ygeno.2018.07.010] [Citation(s) in RCA: 5] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/11/2018] [Revised: 06/25/2018] [Accepted: 07/11/2018] [Indexed: 12/21/2022]
Abstract
Gene- and pathway-based variant association tests are important tools in finding genetic variants that are associated with phenotypes of interest. Although some methods have been proposed in the literature, powerful and robust statistical tests are still desirable in this area. In this study, we propose a statistical test based on decomposing the genotype data into orthogonal parts from which powerful and robust independent p-value combination approaches can be utilized. Through a comprehensive simulation study, we compare the proposed test with some existing popular ones. Our simulation results show that the new test has great performance in terms of controlling type I error rate and statistical power. Real data applications are also conducted to illustrate the performance and usefulness of the proposed test.
Collapse
Affiliation(s)
- Zhongxue Chen
- Department of Epidemiology and Biostatistics, School of Public Health, Indiana University Bloomington, 1025 E. 7th street, Bloomington, IN 47405, USA.
| | - Qingzhong Liu
- Department of Computer Science, Sam Houston State University, 1803 Avenue I, Huntsville, TX 77341, USA
| | - Kai Wang
- Department of Biostatistics, College of Public Health, University of Iowa, 145 N. Riverside Drive, Iowa City, IA 52242, USA
| |
Collapse
|
5
|
Chen Z, Lu Y, Lin T, Liu Q, Wang K. Gene-based genetic association test with adaptive optimal weights. Genet Epidemiol 2017; 42:95-103. [DOI: 10.1002/gepi.22098] [Citation(s) in RCA: 8] [Impact Index Per Article: 1.1] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/19/2017] [Accepted: 10/22/2017] [Indexed: 12/13/2022]
Affiliation(s)
- Zhongxue Chen
- Department of Epidemiology and Biostatistics; School of Public Health; Indiana University Bloomington; Bloomington Indiana United States of America
| | - Yan Lu
- Department of Mathematics and Statistics; University of New Mexico; Albuquerque New Mexico United States of America
| | - Tong Lin
- The Key Laboratory of Machine Perception (Ministry of Education); School of EECS; Peking University; Beijing China
| | - Qingzhong Liu
- Department of Computer Science; Sam Houston State University; Huntsville Texas United States of America
| | - Kai Wang
- Department of Biostatistics; College of Public Health; University of Iowa; Iowa City Iowa United States of America
| |
Collapse
|
6
|
Gaye A, Davis SK. Genetic model misspecification in genetic association studies. BMC Res Notes 2017; 10:569. [PMID: 29115983 PMCID: PMC5678796 DOI: 10.1186/s13104-017-2911-3] [Citation(s) in RCA: 5] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/01/2017] [Accepted: 11/01/2017] [Indexed: 02/08/2023] Open
Abstract
Objective The underlying model of the genetic determinant of a trait is generally not known with certainty a priori. Hence, in genetic association studies, a dominant model might be erroneously modelled as additive, an error investigated previously. We explored this question, for candidate gene studies, by evaluating the sample size required to compensate for the misspecification and improve inference at the analysis stage. Power calculations were carried out with (1) the true dominant model and (2) the incorrect additive model. Empirical power, sample size and effect size were compared between scenarios (1) and (2). In each of the scenarios the estimates were evaluated for a rare (minor allele frequency < 0.01), low frequency (0.01 ≤ minor allele frequency < 0.05) and common (minor allele frequency ≥ 0.05) single nucleotide polymorphism. Results The results confirm the detrimental effect of the misspecification error on power and effect size for any minor allele frequency. The implications of the error are not negligible; therefore, candidate gene studies should consider the more conservative sample size to compensate for the effect of error. When it is not possible to extend the sample size, methods that help mitigate the impact of the error should be systematically used. Electronic supplementary material The online version of this article (10.1186/s13104-017-2911-3) contains supplementary material, which is available to authorized users.
Collapse
Affiliation(s)
- Amadou Gaye
- Metabolic, Cardiovascular and Inflammatory Disease Genomics Branch, Social Epidemiology Research Unit, National Institutes of Health, National Human Genome Research Institute, Bethesda, USA.
| | - Sharon K Davis
- Metabolic, Cardiovascular and Inflammatory Disease Genomics Branch, Social Epidemiology Research Unit, National Institutes of Health, National Human Genome Research Institute, Bethesda, USA
| |
Collapse
|
7
|
A Powerful Variant-Set Association Test Based on Chi-Square Distribution. Genetics 2017; 207:903-910. [PMID: 28912342 DOI: 10.1534/genetics.117.300287] [Citation(s) in RCA: 13] [Impact Index Per Article: 1.9] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/04/2017] [Accepted: 09/10/2017] [Indexed: 01/19/2023] Open
Abstract
Detecting the association between a set of variants and a given phenotype has attracted a large amount of attention in the scientific community, although it is a difficult task. Recently, several related statistical approaches have been proposed in the literature; powerful statistical tests are still highly desired and yet to be developed in this area. In this paper, we propose a powerful test that combines information from each individual single nucleotide polymorphism (SNP) based on principal component analysis without relying on the eigenvalues associated with the principal components. We compare the proposed approach with some popular tests through a simulation study and real data applications. Our results show that, in general, the new test is more powerful than its competitors considered in this study; the gain in detecting power can be substantial in many situations.
Collapse
|
8
|
A gene-based test of association through an orthogonal decomposition of genotype scores. Hum Genet 2017; 136:1385-1394. [PMID: 28864915 DOI: 10.1007/s00439-017-1839-y] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/30/2017] [Accepted: 08/26/2017] [Indexed: 10/18/2022]
Abstract
The burden test and the sequence kernel association test (SKAT) are two popular methods for detecting association with rare variants. Treated as two different sources of association information, they are adaptively combined to form an optimal SKAT (SKAT-O) method for optimal power. We show that the burden test is part of rather than independent of the SKAT. We introduce a new test statistic that is the sum of the burden statistic and a statistic asymptotically independent of the burden statistic. The performance of this new test statistic is demonstrated through extensive simulation studies and applications to a Genetic Analysis Workshop 17 data set and the Ocular Hypertension Treatment Study data.
Collapse
|
9
|
Chen Z, Han S, Wang K. Genetic association test based on principal component analysis. Stat Appl Genet Mol Biol 2017; 16:189-198. [DOI: 10.1515/sagmb-2016-0061] [Citation(s) in RCA: 9] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/14/2022]
Abstract
AbstractMany gene- and pathway-based association tests have been proposed in the literature. Among them, the SKAT is widely used, especially for rare variants association studies. In this paper, we investigate the connection between SKAT and a principal component analysis. This investigation leads to a procedure that encompasses SKAT as a special case. Through simulation studies and real data applications, we compare the proposed method with some existing tests.
Collapse
|
10
|
Xiong W, Su Y, Ding J. Empirical likelihood-based robust tests for genetic association analysis with quantitative traits. J Appl Stat 2016. [DOI: 10.1080/02664763.2016.1266469] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/20/2022]
Affiliation(s)
- Wenjun Xiong
- School of Mathematics and Statistics, Guangxi Normal University, Guilin, People's Republic of China
| | - You Su
- School of Mathematics and Statistics, Guangxi Normal University, Guilin, People's Republic of China
| | - Juan Ding
- School of Mathematics and Statistics, Guangxi Normal University, Guilin, People's Republic of China
- Department of Medicine, Vanderbilt University School of Medicine, Nashville, TN, USA
| |
Collapse
|
11
|
Chen Z, Yang W, Liu Q, Yang JY, Li J, Yang M. A new statistical approach to combining p-values using gamma distribution and its application to genome-wide association study. BMC Bioinformatics 2014; 15 Suppl 17:S3. [PMID: 25559433 PMCID: PMC4304193 DOI: 10.1186/1471-2105-15-s17-s3] [Citation(s) in RCA: 21] [Impact Index Per Article: 2.1] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/04/2023] Open
Abstract
Background Combining information from different studies is an important and useful practice in bioinformatics, including genome-wide association study, rare variant data analysis and other set-based analyses. Many statistical methods have been proposed to combine p-values from independent studies. However, it is known that there is no uniformly most powerful test under all conditions; therefore, finding a powerful test in specific situation is important and desirable. Results In this paper, we propose a new statistical approach to combining p-values based on gamma distribution, which uses the inverse of the p-value as the shape parameter in the gamma distribution. Conclusions Simulation study and real data application demonstrate that the proposed method has good performance under some situations.
Collapse
|
12
|
Chen Z, Ng HKT, Li J, Liu Q, Huang H. Detecting associated single-nucleotide polymorphisms on the X chromosome in case control genome-wide association studies. Stat Methods Med Res 2014; 26:567-582. [PMID: 25253574 DOI: 10.1177/0962280214551815] [Citation(s) in RCA: 23] [Impact Index Per Article: 2.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/21/2022]
Abstract
In the past decade, hundreds of genome-wide association studies have been conducted to detect the significant single-nucleotide polymorphisms that are associated with certain diseases. However, most of the data from the X chromosome were not analyzed and only a few significant associated single-nucleotide polymorphisms from the X chromosome have been identified from genome-wide association studies. This is mainly due to the lack of powerful statistical tests. In this paper, we propose a novel statistical approach that combines the information of single-nucleotide polymorphisms on the X chromosome from both males and females in an efficient way. The proposed approach avoids the need of making strong assumptions about the underlying genetic models. Our proposed statistical test is a robust method that only makes the assumption that the risk allele is the same for both females and males if the single-nucleotide polymorphism is associated with the disease for both genders. Through simulation study and a real data application, we show that the proposed procedure is robust and have excellent performance compared to existing methods. We expect that many more associated single-nucleotide polymorphisms on the X chromosome will be identified if the proposed approach is applied to current available genome-wide association studies data.
Collapse
Affiliation(s)
- Zhongxue Chen
- 1 Department of Epidemiology and Biostatistics, School of Public Health, Indiana University Bloomington, Bloomington, IN, USA
| | - Hon Keung Tony Ng
- 2 Department of Statistical Science, Southern Methodist University, Dallas, TX, USA
| | - Jing Li
- 1 Department of Epidemiology and Biostatistics, School of Public Health, Indiana University Bloomington, Bloomington, IN, USA
| | - Qingzhong Liu
- 3 Department of Computer Science, Sam Houston State Uiversity, Huntsville, TX, USA
| | - Hanwen Huang
- 4 Department of Epidemiology and Biostatistics, College of Public Health, University of Georgia, Athens, GA, USA
| |
Collapse
|
13
|
|
14
|
Chen Z, Huang H, Liu Q. Detecting differentially methylated loci for multiple treatments based on high-throughput methylation data. BMC Bioinformatics 2014; 15:142. [PMID: 24884464 PMCID: PMC4026834 DOI: 10.1186/1471-2105-15-142] [Citation(s) in RCA: 5] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/10/2013] [Accepted: 05/06/2014] [Indexed: 12/31/2022] Open
Abstract
Background Because of its important effects, as an epigenetic factor, on gene expression and disease development, DNA methylation has drawn much attention from researchers. Detecting differentially methylated loci is an important but challenging step in studying the regulatory roles of DNA methylation in a broad range of biological processes and diseases. Several statistical approaches have been proposed to detect significant methylated loci; however, most of them were designed specifically for case-control studies. Results Noticing that the age is associated with methylation level and the methylation data are not normally distributed, in this paper, we propose a nonparametric method to detect differentially methylated loci under multiple conditions with trend for Illumina Array Methylation data. The nonparametric method, Cuzick test is used to detect the differences among treatment groups with trend for each age group; then an overall p-value is calculated based on the method of combining those independent p-values each from one age group. Conclusions We compare the new approach with other methods using simulated and real data. Our study shows that the proposed method outperforms other methods considered in this paper in term of power: it detected more biological meaningful differentially methylated loci than others.
Collapse
Affiliation(s)
- Zhongxue Chen
- Department of Epidemiology and Biostatistics, School of Public Health, Indiana University Bloomington, 1025 E, 7th street, PH C104, Bloomington, IN 47405, USA.
| | | | | |
Collapse
|
15
|
Chen Z. A new association test based on disease allele selection for case-control genome-wide association studies. BMC Genomics 2014; 15:358. [PMID: 24886381 PMCID: PMC4059871 DOI: 10.1186/1471-2164-15-358] [Citation(s) in RCA: 9] [Impact Index Per Article: 0.9] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/05/2014] [Accepted: 05/06/2014] [Indexed: 12/20/2022] Open
Abstract
Background Current robust association tests for case–control genome-wide association study (GWAS) data are mainly based on the assumption of some specific genetic models. Due to the richness of the genetic models, this assumption may not be appropriate. Therefore, robust but powerful association approaches are desirable. Results In this paper, we propose a new approach to testing for the association between the genotype and phenotype for case–control GWAS. This method assumes a generalized genetic model and is based on the selected disease allele to obtain a p-value from the more powerful one-sided test. Through a comprehensive simulation study we assess the performance of the new test by comparing it with existing methods. Some real data applications are also used to illustrate the use of the proposed test. Conclusions Based on the simulation results and real data application, the proposed test is powerful and robust. Electronic supplementary material The online version of this article (doi:10.1186/1471-2164-15-358) contains supplementary material, which is available to authorized users.
Collapse
Affiliation(s)
- Zhongxue Chen
- Department of Epidemiology and Biostatistics, School of Public Health, Indiana University Bloomington, 1025 E, 7th street, PH C104, Bloomington, IN 47405, USA.
| |
Collapse
|
16
|
Chen Z, Nadarajah S. On the optimally weighted z-test for combining probabilities from independent studies. Comput Stat Data Anal 2014. [DOI: 10.1016/j.csda.2013.09.005] [Citation(s) in RCA: 12] [Impact Index Per Article: 1.2] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/22/2022]
|