scholarly journals DeepNull models non-linear covariate effects to improve phenotypic prediction and association power

2022 ◽  
Vol 13 (1) ◽  
Author(s):  
Zachary R. McCaw ◽  
Thomas Colthurst ◽  
Taedong Yun ◽  
Nicholas A. Furlotte ◽  
Andrew Carroll ◽  
...  
Keyword(s):  
Statistical Power ◽  
Association Studies ◽  
Real Data ◽  
Interactive Effects ◽  
Type I ◽  
Genome Wide Association Studies ◽  
Linear Modeling ◽  
Association Analyses ◽  
Covariate Effects ◽  
Non Linear

AbstractGenome-wide association studies (GWASs) examine the association between genotype and phenotype while adjusting for a set of covariates. Although the covariates may have non-linear or interactive effects, due to the challenge of specifying the model, GWAS often neglect such terms. Here we introduce DeepNull, a method that identifies and adjusts for non-linear and interactive covariate effects using a deep neural network. In analyses of simulated and real data, we demonstrate that DeepNull maintains tight control of the type I error while increasing statistical power by up to 20% in the presence of non-linear and interactive effects. Moreover, in the absence of such effects, DeepNull incurs no loss of power. When applied to 10 phenotypes from the UK Biobank (n = 370K), DeepNull discovered more hits (+6%) and loci (+7%), on average, than conventional association analyses, many of which are biologically plausible or have previously been reported. Finally, DeepNull improves upon linear modeling for phenotypic prediction (+23% on average).

scholarly journals DeepNull: Modeling non-linear covariate effects improves phenotype prediction and association power

2021 ◽  
Author(s):  
Farhad Hormozdiari ◽  
Zachary R Mccaw ◽  
Thomas Colthurst ◽  
Taedong Yun ◽  
Nick Furlotte ◽  
...  
Keyword(s):  
Complex Traits ◽  
Statistical Power ◽  
Association Studies ◽  
Simulated Data ◽  
Type I ◽  
Genome Wide Association Studies ◽  
Phenotype Prediction ◽  
Covariate Effects ◽  
Non Linear ◽  
Linear Effects

Genome-wide association studies (GWAS) are among the workhorses of statistical genetics, having detected thousands of variants associated with complex traits and diseases. A typical GWAS examines the association between genotypes and the phenotype of interest while adjusting for a set of covariates. While covariates potentially have non-linear effects on the phenotype in many real world settings, due to the challenge of specifying the model, GWAS seldom include non-linear terms. Here we introduce DeepNull, a method that models non-linear covariate effects on phenotypes using a deep neural network (DNN) and then includes the model prediction as a single extra term in the GWAS association. First, using simulated data, we show that DeepNull increases statistical power by up to 20% while maintaining tight control of the type I error in the presence of interactions or non-linear covariate effects. Second, DeepNull maintains similar results to a standard GWAS when covariates have only linear effects on the phenotype. Third, DeepNull detects larger numbers of significant hits and loci (7% additional loci averaged over 10 traits) than standard GWAS in ten phenotypes from the UK Biobank (n=370K). Many of the hits found only by DeepNull are biologically plausible or have previously been reported in the GWAS catalog. Finally, DeepNull improves phenotype prediction by 23% averaged over the same ten phenotypes, the highest improvement was observed in the case of Glaucoma referral probability where DeepNull improves the phenotype prediction by 83%.

scholarly journals A comprehensive evaluation of methods for Mendelian randomization using realistic simulations and an analysis of 38 biomarkers for risk of type 2 diabetes

2021 ◽  
Author(s):  
Guanghao Qi ◽  
Nilanjan Chatterjee
Keyword(s):  
Type 2 Diabetes ◽  
Mendelian Randomization ◽  
Association Studies ◽  
Real Data ◽  
Causal Effects ◽  
Type I ◽  
Genome Wide Association Studies ◽  
Simulation Studies ◽  
Sample Sizes

Abstract Background Previous studies have often evaluated methods for Mendelian randomization (MR) analysis based on simulations that do not adequately reflect the data-generating mechanisms in genome-wide association studies (GWAS) and there are often discrepancies in the performance of MR methods in simulations and real data sets. Methods We use a simulation framework that generates data on full GWAS for two traits under a realistic model for effect-size distribution coherent with the heritability, co-heritability and polygenicity typically observed for complex traits. We further use recent data generated from GWAS of 38 biomarkers in the UK Biobank and performed down sampling to investigate trends in estimates of causal effects of these biomarkers on the risk of type 2 diabetes (T2D). Results Simulation studies show that weighted mode and MRMix are the only two methods that maintain the correct type I error rate in a diverse set of scenarios. Between the two methods, MRMix tends to be more powerful for larger GWAS whereas the opposite is true for smaller sample sizes. Among the other methods, random-effect IVW (inverse-variance weighted method), MR-Robust and MR-RAPS (robust adjust profile score) tend to perform best in maintaining a low mean-squared error when the InSIDE assumption is satisfied, but can produce large bias when InSIDE is violated. In real-data analysis, some biomarkers showed major heterogeneity in estimates of their causal effects on the risk of T2D across the different methods and estimates from many methods trended in one direction with increasing sample size with patterns similar to those observed in simulation studies. Conclusion The relative performance of different MR methods depends heavily on the sample sizes of the underlying GWAS, the proportion of valid instruments and the validity of the InSIDE assumption. Down-sampling analysis can be used in large GWAS for the possible detection of bias in the MR methods.

scholarly journals Block HSIC Lasso: model-free biomarker detection for ultra-high dimensional data

2019 ◽  
Vol 35 (14) ◽  
pp. i427-i435 ◽  
Author(s):  
Héctor Climente-González ◽  
Chloé-Agathe Azencott ◽  
Samuel Kaski ◽  
Makoto Yamada
Keyword(s):  
Single Cell ◽  
Rna Sequencing ◽  
Association Studies ◽  
Real Data ◽  
Supplementary Information ◽  
Genome Wide Association Studies ◽  
Model Free ◽  
Computational Overhead ◽  
Single Cell Rna Sequencing ◽  
Non Linear

AbstractMotivationFinding non-linear relationships between biomolecules and a biological outcome is computationally expensive and statistically challenging. Existing methods have important drawbacks, including among others lack of parsimony, non-convexity and computational overhead. Here we propose block HSIC Lasso, a non-linear feature selector that does not present the previous drawbacks.ResultsWe compare block HSIC Lasso to other state-of-the-art feature selection techniques in both synthetic and real data, including experiments over three common types of genomic data: gene-expression microarrays, single-cell RNA sequencing and genome-wide association studies. In all cases, we observe that features selected by block HSIC Lasso retain more information about the underlying biology than those selected by other techniques. As a proof of concept, we applied block HSIC Lasso to a single-cell RNA sequencing experiment on mouse hippocampus. We discovered that many genes linked in the past to brain development and function are involved in the biological differences between the types of neurons.Availability and implementationBlock HSIC Lasso is implemented in the Python 2/3 package pyHSICLasso, available on PyPI. Source code is available on GitHub (https://github.com/riken-aip/pyHSICLasso).Supplementary informationSupplementary data are available at Bioinformatics online.

scholarly journals A comparative analysis of cell-type adjustment methods for epigenome-wide association studies based on simulated and real data sets

2018 ◽  
Vol 20 (6) ◽  
pp. 2055-2065 ◽  
Author(s):  
Johannes Brägelmann ◽  
Justo Lorenzo Bermejo
Keyword(s):  
Statistical Power ◽  
Type I Error ◽  
Association Studies ◽  
Real Data ◽  
Error Rates ◽  
Data Sets ◽  
Type I ◽  
Cell Type ◽  
Type I Error Rates

Abstract Technological advances and reduced costs of high-density methylation arrays have led to an increasing number of association studies on the possible relationship between human disease and epigenetic variability. DNA samples from peripheral blood or other tissue types are analyzed in epigenome-wide association studies (EWAS) to detect methylation differences related to a particular phenotype. Since information on the cell-type composition of the sample is generally not available and methylation profiles are cell-type specific, statistical methods have been developed for adjustment of cell-type heterogeneity in EWAS. In this study we systematically compared five popular adjustment methods: the factored spectrally transformed linear mixed model (FaST-LMM-EWASher), the sparse principal component analysis algorithm ReFACTor, surrogate variable analysis (SVA), independent SVA (ISVA) and an optimized version of SVA (SmartSVA). We used real data and applied a multilayered simulation framework to assess the type I error rate, the statistical power and the quality of estimated methylation differences according to major study characteristics. While all five adjustment methods improved false-positive rates compared with unadjusted analyses, FaST-LMM-EWASher resulted in the lowest type I error rate at the expense of low statistical power. SVA efficiently corrected for cell-type heterogeneity in EWAS up to 200 cases and 200 controls, but did not control type I error rates in larger studies. Results based on real data sets confirmed simulation findings with the strongest control of type I error rates by FaST-LMM-EWASher and SmartSVA. Overall, ReFACTor, ISVA and SmartSVA showed the best comparable statistical power, quality of estimated methylation differences and runtime.

scholarly journals A fast mrMLM algorithm for multi-locus genome-wide association studies

2018 ◽  
Author(s):  
Cox Lwaka Tamba ◽  
Yuan-Ming Zhang
Keyword(s):  
False Positive ◽  
Statistical Power ◽  
Association Studies ◽  
False Positive Rate ◽  
Real Data ◽  
High Accuracy ◽  
Genome Wide Association ◽  
Genome Wide Association Studies ◽  
Genome Wide ◽  
Positive Rate

AbstractBackgroundRecent developments in technology result in the generation of big data. In genome-wide association studies (GWAS), we can get tens of million SNPs that need to be tested for association with a trait of interest. Indeed, this poses a great computational challenge. There is a need for developing fast algorithms in GWAS methodologies. These algorithms must ensure high power in QTN detection, high accuracy in QTN estimation and low false positive rate.ResultsHere, we accelerated mrMLM algorithm by using GEMMA idea, matrix transformations and identities. The target functions and derivatives in vector/matrix forms for each marker scanning are transformed into some simple forms that are easy and efficient to evaluate during each optimization step. All potentially associated QTNs with P-values ≤ 0.01 are evaluated in a multi-locus model by LARS algorithm and/or EM-Empirical Bayes. We call the algorithm FASTmrMLM. Numerical simulation studies and real data analysis validated the FASTmrMLM. FASTmrMLM reduces the running time in mrMLM by more than 50%. FASTmrMLM also shows high statistical power in QTN detection, high accuracy in QTN estimation and low false positive rate as compared to GEMMA, FarmCPU and mrMLM. Real data analysis shows that FASTmrMLM was able to detect more previously reported genes than all the other methods: GEMMA/EMMA, FarmCPU and mrMLM.ConclusionsFASTmrMLM is a fast and reliable algorithm in multi-locus GWAS and ensures high statistical power, high accuracy of estimates and low false positive rate.Author SummaryThe current developments in technology result in the generation of a vast amount of data. In genome-wide association studies, we can get tens of million markers that need to be tested for association with a trait of interest. Due to the computational challenge faced, we developed a fast algorithm for genome-wide association studies. Our approach is a two stage method. In the first step, we used matrix transformations and identities to quicken the testing of each random marker effect. The target functions and derivatives which are in vector/matrix forms for each marker scanning are transformed into some simple forms that are easy and efficient to evaluate during each optimization step. In the second step, we selected all potentially associated SNPs and evaluated them in a multi-locus model. From simulation studies, our algorithm significantly reduces the computing time. The new method also shows high statistical power in detecting significant markers, high accuracy in marker effect estimation and low false positive rate. We also used the new method to identify relevant genes in real data analysis. We recommend our approach as a fast and reliable method for carrying out a multi-locus genome-wide association study.

scholarly journals MARS: leveraging allelic heterogeneity to increase power of association testing

2021 ◽  
Vol 22 (1) ◽  
Author(s):  
Farhad Hormozdiari ◽  
Junghyun Jung ◽  
Eleazar Eskin ◽  
Jong Wha J. Joo
Keyword(s):  
Type I Error ◽  
Association Studies ◽  
Simulated Data ◽  
Real Data ◽  
Association Test ◽  
Type I ◽  
Genome Wide Association Studies ◽  
Association Testing ◽  
Causal Status ◽  
Causal Variants

AbstractIn standard genome-wide association studies (GWAS), the standard association test is underpowered to detect associations between loci with multiple causal variants with small effect sizes. We propose a statistical method, Model-based Association test Reflecting causal Status (MARS), that finds associations between variants in risk loci and a phenotype, considering the causal status of variants, only requiring the existing summary statistics to detect associated risk loci. Utilizing extensive simulated data and real data, we show that MARS increases the power of detecting true associated risk loci compared to previous approaches that consider multiple variants, while controlling the type I error.

scholarly journals An Optimally Weighted Combination Method to Detect Novel Disease Associated Genes Using Publicly Available GWAS Summary Data

2019 ◽  
Author(s):  
Jianjun Zhang ◽  
Samantha Gonzales ◽  
Jianguo Liu ◽  
Xiaoyi Raymond Gao ◽  
Xuexia Wang
Keyword(s):  
Statistical Power ◽  
Association Studies ◽  
Score Test ◽  
Real Data ◽  
Test Method ◽  
Combination Method ◽  
Genome Wide Association Studies ◽  
Weighted Sum ◽  
Special Cases ◽  
Weighted Combination

AbstractGene-based analyses offer a useful alternative and complement to the usual single nucleotide polymorphism (SNP) based analysis for genome-wide association studies (GWASs). Using appropriate weights (pre-specified or eQTL-derived) can boost statistical power, especially for detecting weak associations between a gene and a trait. Because the sparsity level or association directions of the underlying association patterns in real data are often unknown and access to individual-level data is limited, we propose an optimal weighted combination (OWC) test applicable to summary statistics from GWAS. This method includes burden tests, weighted sum of squared score (SSU), weighted sum statistic (WSS), and the score test as its special cases. We analytically prove that aggregating the variants in one gene is the same as using the weighted combination of Z-scores for each variant based on the score test method. We also numerically illustrate that our proposed test outperforms several existing comparable methods via simulation studies. Lastly, we utilize schizophrenia GWAS data and a fasting glucose GWAS meta-analysis data to demonstrate that our method outperforms the existing methods in real data analyses. Our proposed test is implemented in the R program OWC, which is freely and publicly available.

scholarly journals Genome-wide Marginal Epistatic Association Mapping in Case-Control Studies

2018 ◽  
Author(s):  
Lorin Crawford ◽  
Xiang Zhou
Keyword(s):  
Complex Traits ◽  
Statistical Power ◽  
Association Studies ◽  
Computational Cost ◽  
Case Control ◽  
Type I ◽  
Genome Wide Association Studies ◽  
Case Control Studies ◽  
Control Data ◽  
Genome Wide

AbstractEpistasis, commonly defined as the interaction between genetic loci, is an important contributor to the genetic architecture underlying many complex traits and common diseases. Most existing epistatic mapping methods in genome-wide association studies explicitly search over all pairwise or higher-order interactions. However, due to the potentially large search space and the resulting multiple testing burden, these conventional approaches often suffer from heavy computational cost and low statistical power. A recently proposed attractive alternative for mapping epistasis focuses instead on detecting marginal epistasis, which is defined as the combined pairwise interaction effects between a given variant and all other variants. By searching for marginal epistatic effects, one can identify genetic variants that are involved in epistasis without the need to identify the exact partners with which the variants interact — thus, potentially alleviating much of the statistical and computational burden associated with conventional epistatic mapping procedures. However, previous marginal epistatic mapping methods are based on quantitative trait models. As we will show here, these lack statistical power in case-control studies. Here, we develop a liability threshold mixed model that extends marginal epistatic mapping to case-control studies. Our method properly accounts for case-control ascertainment and the binary nature of case-control data. We refer to this method as the liability threshold marginal epistasis test (LT-MAPIT). With simulations, we illustrate the benefits of LT-MAPIT in terms of providing effective type I error control, and being more powerful than both existing marginal epistatic mapping methods and conventional explicit search-based approaches in case-control data. We finally apply LT-MAPIT to identify both marginal and pairwise epistasis in seven complex diseases from the Wellcome Trust Case Control Consortium (WTCCC) 1 study.

scholarly journals Leveraging allelic heterogeneity to increase power of association testing

2018 ◽  
Author(s):  
Farhad I Hormozdiari ◽  
Junghyun Jung ◽  
Eleazar Eskin ◽  
Jong Wha J. Joo
Keyword(s):  
Statistical Power ◽  
Association Studies ◽  
Simulated Data ◽  
Association Test ◽  
Type I ◽  
Genome Wide Association Studies ◽  
Summary Statistics ◽  
Causal Status ◽  
Causal Variants ◽  
Multiple Variants

The standard genome-wide association studies (GWAS) detects an association between a single variant and a phenotype of interest. Recently, several studies reported that at many risk loci, there may exist multiple causal variants. For a locus with multiple causal variants with small effect sizes, the standard association test is underpowered to detect the associations. Alternatively, an approach considering effects of multiple variants simultaneously may increase statistical power by leveraging effects of multiple causal variants. In this paper, we propose a new statistical method, Model-based Association test Reflecting causal Status (MARS), that tries to find an association between variants in risk loci and a phenotype, considering the causal status of the variants. One of the main advantages of MARS is that it only requires the existing summary statistics to detect associated risk loci. Thus, MARS is applicable to any association study with summary statistics, even though individual level data is not available for the study. Utilizing extensive simulated data sets, we show that MARS increases the power of detecting true associated risk loci compared to previous approaches that consider multiple variants, while robustly controls the type I error. Applied to data of 44 tissues provided by the Genotype-Tissue Expression (GTEx) consortium, we show that MARS identifies more eGenes compared to previous approaches in most of the tissues; e.g. MARS identified 16% more eGenes than the ones reported by the GTEx consortium. Moreover, applied to Northern Finland Birth Cohort (NFBC) data, we demonstrate that MARS effectively identifies association loci with improved power (56% of more loci found by MARS) inGWAS studies compared to the standard association test.

scholarly journals Patterns of convergence and divergence between bipolar disorder type I and type II: evidence from integrative genomic analyses

2021 ◽  
Author(s):  
Yunqi Huang ◽  
Yunjia Liu ◽  
Yulu Wu ◽  
Yiguo Tang ◽  
Siyi Liu ◽  
...  
Keyword(s):  
Bipolar Disorder ◽  
Association Studies ◽  
Brain Regions ◽  
Type I ◽  
Genome Wide Association Studies ◽  
Type Ii ◽  
Association Analyses ◽  
Convergence And Divergence ◽  
The Difference ◽  
Genetic Structures

Genome-wide association studies (GWAS) analyses have revealed genetic evidence of bipolar disorder (BD), but little is known about genetic structure of BD subtypes. We aimed to investigate genetic overlap and distinction of bipolar type I (BDI) & type II (BDII) by conducting integrative post-GWAS analyses. This study utilized single nucleotide polymorphism (SNP)-level approaches to uncover correlated and distinct genetic loci. Transcriptome-wide association analyses (TWAS) were then approached to pinpoint functional genes expressed in specific brain tissues and blood. Next, we performed cross-phenotype analysis including exploring the potential causal associations between BDI & II and drug responses and comparing the difference of genetic structures among four different psychiatric traits. Our results find SNP-level evidence revealed three genomic loci, SLC25A17, ZNF184 and RPL10AP3 shared by BDI & II, while one locus (i.e., MAD1L1) and significant gene sets involved in calcium channel activity, neural and synapsed signals that distinguished two subtypes. TWAS data implicated different genes effecting BDI & II through expression in specific brain regions (e.g., nucleus accumbens for BDI). Cross-phenotype analyses indicated that BDI & II have different drug response, but share continuous genetic structures with schizophrenia (SCZ) and major depression disorder (MDD), which help fill the gaps left by the dichotomy of mental disorder. These combined evidences illustrate genetic convergence and divergence between BDI & II and provide an underlying biological and trans-diagnostic insight into major psychiatric disorders.

Sign in / Sign up

Export Citation Format

Share Document