Control Of False Positive Rate Research Articles

CoPheScan: phenome-wide association studies accounting for linkage disequilibrium

Phenome-wide association studies (PheWAS) facilitate the discovery of associations between a single genetic variant with multiple phenotypes. For variants which impact a specific protein, this can help identify additional therapeutic indications or on-target side effects of intervening on that protein. However, PheWAS is restricted by an inability to distinguish confounding due to linkage disequilibrium (LD) from true pleiotropy. Here we describe CoPheScan (Coloc adapted Phenome-wide Scan), a Bayesian approach that enables an intuitive and systematic exploration of causal associations while simultaneously addressing LD confounding. We demonstrate its performance through simulation, showing considerably better control of false positive rates than a conventional approach not accounting for LD. We used CoPheScan to perform PheWAS of protein-truncating variants and fine-mapped variants from disease and pQTL studies, in 2275 disease phenotypes from the UK Biobank. Our results identify the complexity of known pleiotropic genes such as APOE, and suggest a new causal role for TGM3 in skin cancer.

Bayesian Adaptive Lasso for the Detection of Differential Item Functioning in Graded Response Models

The detection of differential item functioning (DIF) is important in psychological and behavioral sciences. Standard DIF detection methods perform an item-by-item test iteratively, often assuming that all items except the one under investigation are DIF-free. This article proposes a Bayesian adaptive Lasso method to detect DIF in graded response models (GRMs), where the DIF effects for all items can be identified simultaneously. The multiple-group GRMs are specified, and the possible DIF effects for each item are reparameterized using the increment components. Then, a Bayesian adaptive Lasso procedure is developed for parameter estimation, in which DIF effects can be automatically obtained. Our method is evaluated and compared with the commonly used likelihood ratio test method in a simulation study. The results show that our method can recover most model parameters well and has better control of false positive rates in almost all conditions. An application is presented using data from the National Longitudinal Study of Adolescent to Adult Health (Add Health).

XMAP: Cross-population fine-mapping by leveraging genetic diversity and accounting for confounding bias

Fine-mapping prioritizes risk variants identified by genome-wide association studies (GWASs), serving as a critical step to uncover biological mechanisms underlying complex traits. However, several major challenges still remain for existing fine-mapping methods. First, the strong linkage disequilibrium among variants can limit the statistical power and resolution of fine-mapping. Second, it is computationally expensive to simultaneously search for multiple causal variants. Third, the confounding bias hidden in GWAS summary statistics can produce spurious signals. To address these challenges, we develop a statistical method for cross-population fine-mapping (XMAP) by leveraging genetic diversity and accounting for confounding bias. By using cross-population GWAS summary statistics from global biobanks and genomic consortia, we show that XMAP can achieve greater statistical power, better control of false positive rate, and substantially higher computational efficiency for identifying multiple causal signals, compared to existing methods. Importantly, we show that the output of XMAP can be integrated with single-cell datasets, which greatly improves the interpretation of putative causal variants in their cellular context at single-cell resolution.

CEDA: integrating gene expression data with CRISPR-pooled screen data identifies essential genes with higher expression.

Clustered regularly interspaced short palindromic repeats (CRISPR)-based genetic perturbation screen is a powerful tool to probe gene function. However, experimental noises, especially for the lowly expressed genes, need to be accounted for to maintain proper control of false positive rate. We develop a statistical method, named CRISPR screen with Expression Data Analysis (CEDA), to integrate gene expression profiles and CRISPR screen data for identifying essential genes. CEDA stratifies genes based on expression level and adopts a three-component mixture model for the log-fold change of single-guide RNAs (sgRNAs). Empirical Bayesian prior and expectation-maximization algorithm are used for parameter estimation and false discovery rate inference. Taking advantage of gene expression data, CEDA identifies essential genes with higher expression. Compared to existing methods, CEDA shows comparable reliability but higher sensitivity in detecting essential genes with moderate sgRNA fold change. Therefore, using the same CRISPR data, CEDA generates an additional hit gene list. Supplementary data are available at Bioinformatics online.

A Markov random field model-based approach for differentially expressed gene detection from single-cell RNA-seq data.

The development of single-cell RNA-sequencing (scRNA-seq) technologies has offered insights into complex biological systems at the single-cell resolution. In particular, these techniques facilitate the identifications of genes showing cell-type-specific differential expressions (DE). In this paper, we introduce MARBLES, a novel statistical model for cross-condition DE gene detection from scRNA-seq data. MARBLES employs a Markov Random Field model to borrow information across similar cell types and utilizes cell-type-specific pseudobulk count to account for sample-level variability. Our simulation results showed that MARBLES is more powerful than existing methods to detect DE genes with an appropriate control of false positive rate. Applications of MARBLES to real data identified novel disease-related DE genes and biological pathways from both a single-cell lipopolysaccharide mouse dataset with 24 381 cells and 11 076 genes and a Parkinson’s disease human data set with 76 212 cells and 15 891 genes. Overall, MARBLES is a powerful tool to identify cell-type-specific DE genes across conditions from scRNA-seq data.

Visual Field Progression in Glaucoma: Comparison Between PoPLR and ANSWERS

PurposeIt has been suggested that the detection of visual field progression can be improved by modeling statistical properties of the data such as the increasing retest variability and the spatial correlation among visual field locations. We compared a method that models those properties, Analysis with Non-Stationary Weibull Error Regression and Spatial Enhancement (ANSWERS), against a simpler one that does not, Permutation of Pointwise Linear Regression (PoPLR).MethodsVisual field series from three independent longitudinal studies in patients with glaucoma were used to compare the positive rate of PoPLR and ANSWERS. To estimate the false-positive rate, the same visual field series were randomly re-ordered in time. The first dataset consisted of series of 7 visual fields from 101 eyes, the second consisted of series of 9 visual fields from 150 eyes, and the third consisted of series of more than 9 visual fields (17.5 on average) from 139 eyes.ResultsFor a statistical significance of 0.05, the false-positive rates for ANSWERS were about 3 times greater than expected at 15%, 17%, and 16%, respectively, whereas for PoPLR they were 7%, 3%, and 6%. After equating the specificities at 0.05 for both models, positive rates for ANSWERS were 16%, 25%, and 38%, whereas for PoPLR they were 12%, 33%, and 49%, or about 5% greater on average (95% confidence interval = −1% to 11%).ConclusionsDespite being simpler and less computationally demanding, PoPLR was at least as sensitive to deterioration as ANSWERS once the specificities were equated.Translational RelevanceClose control of false-positive rates is key when visual fields of patients are analyzed for change in both clinical practice and clinical trials.

The power of word-frequency-based alignment-free functions: a comprehensive large-scale experimental analysis.

Alignment-free (AF) distance/similarity functions are a key tool for sequence analysis. Experimental studies on real datasets abound and, to some extent, there are also studies regarding their control of false positive rate (Type I error). However, assessment of their power, i.e. their ability to identify true similarity, has been limited to some members of the D2 family. The corresponding experimental studies have concentrated on short sequences, a scenario no longer adequate for current applications, where sequence lengths may vary considerably. Such a State of the Art is methodologically problematic, since information regarding a key feature such as power is either missing or limited. By concentrating on a representative set of word-frequency-based AF functions, we perform the first coherent and uniform evaluation of the power, involving also Type I error for completeness. Two alternative models of important genomic features (CIS Regulatory Modules and Horizontal Gene Transfer), a wide range of sequence lengths from a few thousand to millions, and different values of k have been used. As a result, we provide a characterization of those AF functions that is novel and informative. Indeed, we identify weak and strong points of each function considered, which may be used as a guide to choose one for analysis tasks. Remarkably, of the 15 functions that we have considered, only four stand out, with small differences between small and short sequence length scenarios. Finally, to encourage the use of our methodology for validation of future AF functions, the Big Data platform supporting it is public. The software is available at: https://github.com/pipp8/power_statistics. Supplementary data are available at Bioinformatics online.

Treatment Efficacy Score (TES), a continuous residual cancer burden-based metric to compare neoadjuvant chemotherapy efficacy between trial arms in the I-SPY 2 trial.

587 Background: Residual cancer burden (RCB) is a continuous score that captures the amount of residual cancer after neoadjuvant chemotherapy and predicts disease recurrence and survival across all breast cancer subtypes. RCB score 0 corresponds to pathological complete response (pCR; ypT0, ypN0). We hypothesize that comparison of the distributions of RCB scores between randomized treatment arms of a trial could predict treatment effect on recurrence free survival better than comparison of pCR rates only. Methods: The cancer Treatment Efficacy Score (TES) compares efficacies of two treatments using non-continuous RCB results. We examined (i) area between cumulative distribution (ABC) functions; (ii) density ratio of RCB scores; and (iii) density difference of RCB scores from two treatments, to select the most efficient metric to compute TES. A random permutation procedure was used to estimate the p-value from each test. These methods were applied to data from the durvalumab/olaparib arm and corresponding controls of the I-SPY2 trial, separately by molecular subtype. In subsampling and simulation experiments we assessed robustness of results including power and false positive rate control under variable sample sizes to select the most robust TES metric. The other 11 experimental arms of I-SPY2 were used to assess the performance of the final metric. We calculated correlation between TES and (i) pCR rate difference, and 3- and 5-year (ii) event-free (EFS) and (iii) distant recurrence free survivals (DRFS). Results: RCB scores are multimodal and do not follow normal distribution.In simulated data ABC provided more stable results than the other methods, had good power, performed well with small sample sizes, resulted in low false positive rate, required the least computational time, and therfore was selected as the TES metric for validation in 11 arms of I-SPY2. We found a high correlation between difference in pCR rate and TES value across all molecular subtypes in each of the 11 trial arms (r = 0.92, p = 1.7e-8). There was also significant linear relationship between TES and survival estimates in EFS (r = 0.58, p = 9.3e-3 for 3-years survival; r = 0.62, p = 4.8e-3 for 5-years survival) and DRFS (r = 0.56, p = 1.2e-2 for 3-years survival; r = 0.54, p = 1.8e-2 for 5-years survival). Statistically significant TES score correlated significantly with higher benefit in 3-years survival (p = 9.7e-4 for EFS; p = 5.7e-3 for DRFS) and 5-years survival (p = 9.7e-4 for EFS; p = 3.0e-3 for DRFS). In most instances, this correlation with survival was higher than seen with pCR difference. Conclusions: TES is a novel more optimal metric to identify the more effective cytotoxic neoadjuvant regimen from the entire distribution of pathologic response that significantly correlates with event and recurrence free survival and may serve as a better surrogate than pCR rate difference.

A forward selection algorithm to identify mutually exclusive alterations in cancer studies.

Mutual exclusivity analyses provide an effective tool to identify driver genes from passenger genes for cancer studies. Various algorithms have been developed for the detection of mutual exclusivity, but controlling false positive and improving accuracy remain challenging. We propose a forward selection algorithm for identification of mutually exclusive gene sets (FSME) in this paper. The method includes an initial search of seed pair of mutually exclusive (ME) genes and subsequently including more genes into the current ME set. Simulations demonstrated that, compared to recently published approaches (i.e., CoMEt, WExT, and MEGSA), FSME could provide higher precision or recall rate to identify ME gene sets, and had superior control of false positive rates. With application to TCGA real data sets for AML, BRCA, and GBM, we confirmed that FSME can be utilized to discover cancer driver genes.

Enhancing rare variant interpretation in inherited arrhythmias through quantitative analysis of 5182 cases from long QT syndrome and Brugada syndrome consortia cohorts and gnomAD population controls

Abstract Background/Introduction Guidelines for variant interpretation in Mendelian disease set stringent criteria to report a variant as (likely) pathogenic, prioritising control of false positive rate over test sensitivity and diagnostic yield, and require customisation for the specific genetic characteristics of gene-disease dyads. Inherited arrhythmias like long QT syndrome (LQTS) and Brugada syndrome (BrS) are genetically heterogeneous, with missense variants constituting the preponderance of disease-causing variants. Evidence from family segregation or functional assays to confirm pathogenicity are often unavailable or impractical in clinical genetic testing, leading to high rates of variants of uncertain significance and false negative test results. Methods We compared rare variant frequencies from 1847 LQTS (KCNQ1, KCNH2, SCN5A) and 3335 BrS (SCN5A) cases from the International LQTS/BrS Genetics Consortia to population-specific gnomAD data. We propose disease-specific criteria for ACMG/AMP evidence classes – rarity (PM2/BS1 rules) and enrichment of individual (PS4) and domain-specific (PM1) variants in cases over controls. Results Rare SCN5A variant prevalence differed between BrS cases with spontaneous (28.7%) versus induced (15.8%) type 1 Brugada ECG (p=1.3x10–13) and between European (20.8%) and Japanese (8.9%) patients (p=8.8x10–18). Transmembrane regions and specific N-terminus (KCNH2) and C-terminus (KCNQ1/KCNH2) domains were characterised by high enrichment of case variants and >95% probability of pathogenicity. Applying the customised rules, 17.5% of European BrS cases and 73.7% of European LQTS cases had variants classified as (likely) pathogenic, compared to estimated diagnostic yields (case excess over gnomAD) of 19.3%/82.6%. Conclusions Large case/control datasets enable quantitative implementation of ACMG/AMP guidelines and increased sensitivity for inherited arrhythmia genetic testing. Classification of Brugada/LQTS variants Funding Acknowledgement Type of funding source: Foundation. Main funding source(s): Dutch Heart Foundation, Netherlands Organisation for Scientific Research

No Structural Differences Are Revealed by VBM in 'De Novo' Parkinsonian Patients.

The identification of brain morphological alterations in newly diagnosed PD patients (i.e. 'de novo') could potentially serve as a biomarker and accelerate diagnosis. However, presently no consensus exists in the literature possibly due to several factors: small size cohorts, differences in segmentation techniques or bad control of false positive rates. In this study, we use the CAT12 pipeline, to seek for morphological brain differences in gray and white matter of 66 controls and 144 de novo PD patients from the PPMI database. Moreover, we search for subcortical structure differences using the VolBrain pipeline. We found no structural brain differences in this de novo Parkinsonian population, neither in tissues using a whole brain analysis nor in any of nine subcortical structures analyzed separately. We conclude that some results published in the literature may appear as false positives and we contest their reproductibility.

An implementation of N-way repeated measures ANOVA: Effect coding, automated unpacking of interactions, and randomization testing

The paper presents the details of an implementation of repeated measures ANOVA, consisting of a set of functions to organize data and represent contrasts to be tested and run statistical tests. The implementation is focused on uses common in experimental psychology. An arbitrary number of within-subject factors, each with an arbitrary number of levels, can be used. A non-parametric, randomization- and permutation-based formulation of repeated measures ANOVA was defined and implemented. Methods for testing interactions with categorical and continuous between-subject variables are implemented. Post-hoc tests for exploring interactions are automated. Simulations indicate correct control of false positive rate for all types of test. The software provides output with statistics including p-values and partial eta squared.-An open source implementation of repeated measures ANOVA based on effect coding.-Generates p-values and automatized unpacking of interactions for N-factor designs.-A non-parametric test is defined based on permutation tests.

False positive rate control for positive unlabeled learning

Learning classifiers with false positive rate control have drawn intensive attention in applications over past years. While various supervised algorithms have been developed for obtaining low false positive rates, they commonly require the coexistence of both positive and negative samples in data. However, the scenario studied in positive unlabeled (PU) learning is more pervasive in practice. Namely, at inception, most of the data may not have known labels, and the data with known labels may only represent one type of samples. To tackle this challenge, in this paper we propose a new positive unlabeled learning classifier with false positive rate control. In particular, we first prove that in this context employing oft-adopted convex surrogate loss functions, such as the hinge loss function, begets a redundant penalty for false positive rates. Then, we present that the non-convex ramp loss surrogate function can overcome this barrier and show a concave-convex procedure can solve the associated non-convex optimization problem. Finally, we demonstrate the effectiveness of the proposed method through extensive experiments on multiple datasets.

Control of false positive rates in clusterwise fMRI inferences

ABSTRACTRandom field theory (RFT) provided a theoretical foundation for cluster-extent-based thresholding, the most widely used method for multiple comparison correction of statistical maps in neuroimaging research. However, several studies questioned the validity of the standard clusterwise inference in fMRI analyses and observed inflated false positive rates. In particular, Eklund et al. [Cluster failure: Why fMRI inferences for spatial extent have inflated false-positive rates, Proc. Natl. Acad. Sci. 113 (2016), pp. 7900–7905. Available at http://www.pnas.org/content/113/28/7900.abstract] used resting-state fMRI as null data and found false positive rates of up to , which immediately led to many discussions. In this study, we summarize the assumptions in RFT clusterwise inference and propose new parametric ways to approximate the distribution of the cluster size by properly combining the limiting distribution of the cluster size given by Nosko [Local structure of Gaussian random fields in the vicinity of high-level shines, Sov. Math. Dokl. 10 (1969), pp. 1481–1484] and the expected value of the cluster size provided by Friston et al. [Assessing the significance of focal activations using their spatial extent, Hum. Brain Mapp. 1 (1994), pp. 210–220. Available at http://dx.doi.org/10.1002/hbm.460010306]. We evaluated our proposed method using four different classic simulation settings in published papers. Results show that our method produces a more stringent estimation of cluster extent size, which leads to a better control of false positive rates.

Applying Logistic Regression to Detect Differential Item Functioning in Multidimensional Data.

Conventional differential item functioning (DIF) approaches such as logistic regression (LR) often assume unidimensionality of a scale and match participants in the reference and focal groups based on total scores. However, many educational and psychological assessments are multidimensional by design, and a matching variable using total scores that does not reflect the test structure may not be good practice in multidimensional items for DIF detection. We propose the use of all subscores of a scale in LR and compare its performance with alternative matching methods, including the use of total score and individual subscores. We focused on uniform DIF situation in which 250, 500, or 1,000 participants in each group answered 21 items reflecting two dimensions, and the 21st item was the studied item. Five factors were manipulated in the study: (a) the test structure, (b) numbers of cross-loaded items, (c) group differences in latent abilities, (d) the magnitude of DIF, and (e) group sample size. The results showed that, when the studied item measured a single domain, the conventional LR incorporating total scores as a matching variable yielded inflated false positive rates (FPRs) when two groups differed in one latent ability. The situation worsened when one group had a higher ability in one domain and lower ability in another. The LR using a single subscore as the matching variable performed well in terms of FPRs and true positive rates (TPRs) when two groups did not differ in either one latent ability or differed in one latent ability. However, this approach yielded inflated FPRs when two groups differed in two latent abilities. The proposed LR using two subscores yielded well-controlled FPRs across all conditions and yielded the highest TPRs. When the studied item measured two domains, the use of either the total score or two subscores worked well in the control of FPRs and yielded similar TPRs across conditions, whereas the use of a single subscore resulted in inflated FPRs when two groups differed in one or two latent abilities. In conclusion, we recommend the use of multiple subscores to match subjects in DIF detection for multidimensional data.

Performance of epistasis detection methods in semi-simulated GWAS

BackgroundPart of the missing heritability in Genome Wide Association Studies (GWAS) is expected to be explained by interactions between genetic variants, also called epistasis. Various statistical methods have been developed to detect epistasis in case-control GWAS. These methods face major statistical challenges due to the number of tests required, the complexity of the Linkage Disequilibrium (LD) structure, and the lack of consensus regarding the definition of epistasis. Their limited impact in terms of uncovering new biological knowledge might be explained in part by the limited amount of experimental data available to validate their statistical performances in a realistic GWAS context. In this paper, we introduce a simulation pipeline for generating real scale GWAS data, including epistasis and realistic LD structure. We evaluate five exhaustive bivariate interaction methods, fastepi, GBOOST, SHEsisEpi, DSS, and IndOR. Two hundred thirty four different disease scenarios are considered in extensive simulations. We report the performances of each method in terms of false positive rate control, power, area under the ROC curve (AUC), and computation time using a GPU. Finally we compare the result of each methods on a real GWAS of type 2 diabetes from the Welcome Trust Case Control Consortium.ResultsGBOOST, SHEsisEpi and DSS allow a satisfactory control of the false positive rate. fastepi and IndOR present an increase in false positive rate in presence of LD between causal SNPs, with our definition of epistasis. DSS performs best in terms of power and AUC in most scenarios with no or weak LD between causal SNPs. All methods can exhaustively analyze a GWAS with 6.105 SNPs and 15,000 samples in a couple of hours using a GPU.ConclusionThis study confirms that computation time is no longer a limiting factor for performing an exhaustive search of epistasis in large GWAS. For this task, using DSS on SNP pairs with limited LD seems to be a good strategy to achieve the best statistical performance. A combination approach using both DSS and GBOOST is supported by the simulation results and the analysis of the WTCCC dataset demonstrated that this approach can detect distinct genes in epistasis. Finally, weak epistasis between common variants will be detectable with existing methods when GWAS of a few tens of thousands cases and controls are available.

Improving mass-univariate analysis of neuroimaging data by modelling important unknown covariates: Application to Epigenome-Wide Association Studies

Statistical inference on neuroimaging data is often conducted using a mass-univariate model, equivalent to fitting a linear model at every voxel with a known set of covariates. Due to the large number of linear models, it is challenging to check if the selection of covariates is appropriate and to modify this selection adequately. The use of standard diagnostics, such as residual plotting, is clearly not practical for neuroimaging data. However, the selection of covariates is crucial for linear regression to ensure valid statistical inference. In particular, the mean model of regression needs to be reasonably well specified. Unfortunately, this issue is often overlooked in the field of neuroimaging. This study aims to adopt the existing Confounder Adjusted Testing and Estimation (CATE) approach and to extend it for use with neuroimaging data. We propose a modification of CATE that can yield valid statistical inferences using Principal Component Analysis (PCA) estimators instead of Maximum Likelihood (ML) estimators. We then propose a non-parametric hypothesis testing procedure that can improve upon parametric testing. Monte Carlo simulations show that the modification of CATE allows for more accurate modelling of neuroimaging data and can in turn yield a better control of False Positive Rate (FPR) and Family-Wise Error Rate (FWER). We demonstrate its application to an Epigenome-Wide Association Study (EWAS) on neonatal brain imaging and umbilical cord DNA methylation data obtained as part of a longitudinal cohort study. Software for this CATE study is freely available at http://www.bioeng.nus.edu.sg/cfa/Imaging_Genetics2.html.

Linnorm: improved statistical analysis for single cell RNA-seq expression data.

Linnorm is a novel normalization and transformation method for the analysis of single cell RNA sequencing (scRNA-seq) data. Linnorm is developed to remove technical noises and simultaneously preserve biological variations in scRNA-seq data, such that existing statistical methods can be improved. Using real scRNA-seq data, we compared Linnorm with existing normalization methods, including NODES, SAMstrt, SCnorm, scran, DESeq and TMM. Linnorm shows advantages in speed, technical noise removal and preservation of cell heterogeneity, which can improve existing methods in the discovery of novel subtypes, pseudo-temporal ordering of cells, clustering analysis, etc. Linnorm also performs better than existing DEG analysis methods, including BASiCS, NODES, SAMstrt, Seurat and DESeq2, in false positive rate control and accuracy.

Modified Logistic Regression Approaches to Eliminating the Impact of Response Styles on DIF Detection in Likert-Type Scales.

Extreme response styles (ERS) is prevalent in Likert- or rating-type data but previous research has not well-addressed their impact on differential item functioning (DIF) assessments. This study aimed to fill in the knowledge gap and examined their influence on the performances of logistic regression (LR) approaches in DIF detections, including the ordinal logistic regression (OLR) and the logistic discriminant functional analysis (LDFA). Results indicated that both the standard OLR and LDFA yielded severely inflated false positive rates as the magnitude of the differences in ERS increased between two groups. This study proposed a class of modified LR approaches to eliminating the ERS effect on DIF assessment. These proposed modifications showed satisfactory control of false positive rates when no DIF items existed and yielded a better control of false positive rates and more accurate true positive rates under DIF conditions than the conventional LR approaches did. In conclusion, the proposed modifications are recommended in survey research when there are multiple group or cultural groups.

Scan statistics on Poisson random fields with applications in genomics

The detection of local genomic signals using high-throughput DNA sequencing data can be cast as a problem of scanning a Poisson random field for local changes in the rate of the process. We propose a likelihood-based framework for such scans, and derive formulas for false positive rate control and power calculations. The framework can also accommodate modified processes that involve overdispersion. As a specific, detailed example, we consider the detection of insertions and deletions by paired-end DNA-sequencing. We propose several statistics for this problem, compare their power under current experimental designs, and illustrate their application on an Illumina Platinum Genomes data set.