FDR Control Research Articles

ScaDA: A novel statistical method for differential analysis of single-cell chromatin accessibility sequencing data.

Single-cell ATAC-seq sequencing data (scATAC-seq) has been widely used to investigate chromatin accessibility on the single-cell level. One important application of scATAC-seq data analysis is differential chromatin accessibility (DA) analysis. However, the data characteristics of scATAC-seq such as excessive zeros and large variability of chromatin accessibility across cells impose a unique challenge for DA analysis. Existing statistical methods focus on detecting the mean difference of the chromatin accessible regions while overlooking the distribution difference. Motivated by real data exploration that distribution difference exists among cell types, we introduce a novel composite statistical test named "scaDA", which is based on zero-inflated negative binomial model (ZINB), for performing differential distribution analysis of chromatin accessibility by jointly testing the abundance, prevalence and dispersion simultaneously. Benefiting from both dispersion shrinkage and iterative refinement of mean and prevalence parameter estimates, scaDA demonstrates its superiority to both ZINB-based likelihood ratio tests and published methods by achieving the highest power and best FDR control in a comprehensive simulation study. In addition to demonstrating the highest power in three real sc-multiome data analyses, scaDA successfully identifies differentially accessible regions in microglia from sc-multiome data for an Alzheimer's disease (AD) study that are most enriched in GO terms related to neurogenesis and the clinical phenotype of AD, and AD-associated GWAS SNPs.

An Efficient Testing Procedure for High-Dimensional Mediators with FDR Control

An Efficient Testing Procedure for High-Dimensional Mediators with FDR Control

Information-incorporated gene network construction with FDR control.

Large-scale gene expression studies allow gene network construction to uncover associations among genes. To study direct associations among genes, partial correlation-based networks are preferred over marginal correlations. However, FDR control for partial correlation-based network construction is not well-studied. In addition, currently available partial correlation-based methods cannot take existing biological knowledge to help network construction while controlling FDR. In this paper, we propose a method called Partial Correlation Graph with Information Incorporation (PCGII). PCGII estimates partial correlations between each pair of genes by regularized node-wise regression that can incorporate prior knowledge while controlling the effects of all other genes. It handles high-dimensional data where the number of genes can be much larger than the sample size and controls FDR at the same time. We compare PCGII with several existing approaches through extensive simulation studies and demonstrate that PCGII has better FDR control and higher power. We apply PCGII to a plant gene expression dataset where it recovers confirmed regulatory relationships and a hub node, as well as several direct associations that shed light on potential functional relationships in the system. We also introduce a method to supplement observed data with a pseudogene to apply PCGII when no prior information is available, which also allows checking FDR control and power for real data analysis. R package is freely available for download at https://cran.r-project.org/package=PCGII.

Trade-off between predictive performance and FDR control for high-dimensional Gaussian model selection

Trade-off between predictive performance and FDR control for high-dimensional Gaussian model selection

42 White matter injury is driven by HIV duration, immune and vascular factors in virally suppressed people living with HIV: a longitudinal diffusion imaging study

Objective:The aim of the current longitudinal study was to use improved brain white matter integrity outcomes (better at resolving white matter complexity, hence with improved biological significance compared to traditional diffusion tensor imaging - DTI outcomes) while considering baseline age, cardiovascular diseases (CVD), and HIV disease markers impacts on the health of major white matter tracts in virally suppressed people Living with HIV infection (PLHIV) versus demographically, geographically, and life-style comparable HIVnegative controls. Furthermore, white-matter hyperintensity (WMH) and normal-appearing white matter (NAWM) volumes and microstructure were considered.Participants and Methods:At baseline 48 HIV-controls and 84 virally suppressed PLHIV (mean age 55), and at 24-month follow-up, 40 HIV-controls and 75 virally suppressed PLHIV underwent an MRI scan (3T Phillips) collecting a high-resolution anatomical MRI, FLAIR, and a 32-direction diffusion imaging. The diffusion data were processed using mrtrix and intra-cranial volume-corrected outcomes included fibre density (FD), fibre cross-section (log was used; logFC) and a composite fibre density and cross-section (FDC). The volumetric data was first processed in Freesurfer 6.0, and WMH were segmented using the “pgs” U-Net neural network. Using mixed models, we examine the longitudinal mrtrix outcomes across major white matter tracts by HIV status, and associations with CVD (sum of the scaled scores of total cholesterol, HDL, Systolic BP, current smoking, and diabetes) and HIV disease (HIV duration, historical AIDS, nadir CD4, baseline CD4) markers. Additionally, we assessed the volume, and FDC in the periventricular and deep WMH, as well as NAWM, and the associations with CVD and HIV disease markers. We used FDR control procedure (alpha = 0.05), and all p-values reported are FDR adjusted.Results:Relative to controls, PLHIV showed significant reductions (p<.05 - p<.01) of FC, and FDC to a lesser extent, in multiple long cortical association tracts, and within striatal- and thalamic-frontoparietal connections. A small HIV by age interaction was only detected for FC of inferior longitudinal fasciculus (Beta = -0.004, SE = 0.002 p<.04). However, HIV duration (corrected for baseline age) was associated with worse FDC across multiple tracts (p<03 -p<.001). Baseline CD4 counts associated with lower FD in frontal association tracts (p<.05 -p<.005). Furthermore, WMH increased in size with time, age, and higher CVD risk factors, but not HIV status. In PLHIV, deep WMH and NAWM microstructure were both associated with worse CVD but not HIV disease markers.Conclusions:The fine integrity of major white matter tracts is impacted by HIV status, HIV duration and baseline CD4, whereas WMH and NAWM volumes and microstructure are affected by CVD. Our study provides further evidence of the immuno-vascular underpinning of HIV neuropathogenesis in virally suppressed PLHIV. The convergence of these effects in aging PLHIV may lead to early neurodegeneration. Hence, improving CVD health and maintaining high CD4 is critical.

FDR control and power analysis for high-dimensional logistic regression via StabKoff

FDR control and power analysis for high-dimensional logistic regression via StabKoff

A general framework for powerful confounder adjustment in omics association studies.

Genomic data are subject to various sources of confounding, such as demographic variables, biological heterogeneity, and batch effects. To identify genomic features associated with a variable of interest in the presence of confounders, the traditional approach involves fitting a confounder-adjusted regression model to each genomic feature, followed by multiplicity correction. This study shows that the traditional approach is suboptimal and proposes a new two-dimensional false discovery rate control framework (2DFDR+) that provides significant power improvement over the conventional method and applies to a wide range of settings. 2DFDR+ uses marginal independence test statistics as auxiliary information to filter out less promising features, and FDR control is performed based on conditional independence test statistics in the remaining features. 2DFDR+ provides (asymptotically) valid inference from samples in settings where the conditional distribution of the genomic variables given the covariate of interest and the confounders is arbitrary and completely unknown. Promising finite sample performance is demonstrated via extensive simulations and real data applications. R codes and vignettes are available at https://github.com/asmita112358/tdfdr.np.

Change-detection-assisted multiple testing for spatiotemporal data

This paper considers a large-scale multiple testing problem for spatiotemporal data with multiple change points. A data-driven procedure that aims to fully utilize the clustering information is proposed. Specifically, we first develop a new change-point detection algorithm that integrates the kernel-based aggregation of spatial observations with a global loss function at the temporal level to group data into several sets, and then derive an FDR control scheme for set-wise multiple testing. Under some mild conditions on the spatiotemporal dependence structure, FDR is shown to be strongly controlled. Theoretical analysis and numerical studies demonstrate the advantages of the algorithm over competing methods.

Quantitative proteomics I.: Concept, design, and planning of quantitative proteomics experiments.

Quantitative proteomics I.: Concept, design, and planning of quantitative proteomics experiments.

Improving power in functional magnetic resonance imaging by moving beyond cluster-level inference

Inference in neuroimaging typically occurs at the level of focal brain areas or circuits. Yet, increasingly, well-powered studies paint a much richer picture of broad-scale effects distributed throughout the brain, suggesting that many focal reports may only reflect the tip of the iceberg of underlying effects. How focal versus broad-scale perspectives influence the inferences we make has not yet been comprehensively evaluated using real data. Here, we compare sensitivity and specificity across procedures representing multiple levels of inference using an empirical benchmarking procedure that resamples task-based connectomes from the Human Connectome Project dataset (∼1,000 subjects, 7 tasks, 3 resampling group sizes, 7 inferential procedures). Only broad-scale (network and whole brain) procedures obtained the traditional 80% statistical power level to detect an average effect, reflecting >20% more statistical power than focal (edge and cluster) procedures. Power also increased substantially for false discovery rate- compared with familywise error rate-controlling procedures. The downsides are fairly limited; the loss in specificity for broad-scale and FDR procedures was relatively modest compared to the gains in power. Furthermore, the broad-scale methods we introduce are simple, fast, and easy to use, providing a straightforward starting point for researchers. This also points to the promise of more sophisticated broad-scale methods for not only functional connectivity but also related fields, including task-based activation. Altogether, this work demonstrates that shifting the scale of inference and choosing FDR control are both immediately attainable and can help remedy the issues with statistical power plaguing typical studies in the field.

Fcfdr: an R package to leverage continuous and binary functional genomic data in GWAS

BackgroundGenome-wide association studies (GWAS) are limited in power to detect associations that exceed the stringent genome-wide significance threshold. This limitation can be alleviated by leveraging relevant auxiliary data, such as functional genomic data. Frameworks utilising the conditional false discovery rate have been developed for this purpose, and have been shown to increase power for GWAS discovery whilst controlling the false discovery rate. However, the methods are currently only applicable for continuous auxiliary data and cannot be used to leverage auxiliary data with a binary representation, such as whether SNPs are synonymous or non-synonymous, or whether they reside in regions of the genome with specific activity states.ResultsWe describe an extension to the cFDR framework for binary auxiliary data, called “Binary cFDR”. We demonstrate FDR control of our method using detailed simulations, and show that Binary cFDR performs better than a comparator method in terms of sensitivity and FDR control. We introduce an all-encompassing user-oriented CRAN R package (https://annahutch.github.io/fcfdr/; https://cran.r-project.org/web/packages/fcfdr/index.html) and demonstrate its utility in an application to type 1 diabetes, where we identify additional genetic associations.ConclusionsOur all-encompassing R package, fcfdr, serves as a comprehensive toolkit to unite GWAS and functional genomic data in order to increase statistical power to detect genetic associations.

Sequential common rate decrease detection, isolation, and estimation in multiple Poisson processes

ABSTRACT We consider the detection of a common change when M independent Poisson processes are monitored simultaneously where only a portion of the processes have rate decrease after the change time. In contrast to the detection of rate increase (Wu and Wu (2022)), the individual CUSUM and Shiryayev–Roberts (S–R) processes need to be calculated continuously in parallel. A combined CUSUM-SR procedure is proposed by using the sum of S–R processes as the detection process for a common change and after the detection, the M individual CUSUM processes are used to isolate the changed panels with FDR control. The medians of change time estimates from the individual CUSUM or S–R processes from the isolated changed panels are then used to estimate the common change time. Coal mining disaster data in USA are used for illustration.

In-Depth In Vivo Crosslinking in Minutes by a Compact, Membrane-Permeable, and Alkynyl-Enrichable Crosslinker.

Chemical crosslinking coupled with mass spectrometry (CXMS) has emerged as a powerful technique to obtain the dynamic conformations and interaction interfaces of protein complexes. Limited by the poor cell membrane permeability, chemical reactivity, and biocompatibility of crosslinkers, in vivo crosslinking to capture the dynamics of protein complexes with finer temporal resolution and higher coverage is attractive but challenging. In this work, a trifunctional crosslinker bis(succinimidyl) with propargyl tag (BSP), involving compact size, proper amphipathy, and enrichment capacity, was developed to enable better cell membrane permeability and efficient crosslinking in 5 min without obvious cellular interference. Followed by a two-step enrichment method based on click chemistry at the peptide level, 13,098 crosslinked peptides (5068 inter-crosslinked peptides and 8030 intra-crosslinked peptides) were identified under the data threshold of peptide-spectrum matches (PSMs) ≥2 on the basic of the FDR control of 1%, which was the most comprehensive dataset for homo species cells by a non-cleavable crosslinker. Besides, the interactome network comprising 1519 proteins connected by 2913 interaction edges in various intracellular compartments, as well as 80S ribosome structural dynamics, were characterized, showing the great potential of our in vivo crosslinking approach in minutes. All these results demonstrated that our developed BSP could provide a valuable toolkit for the in-depth in vivo analysis of protein-protein interactions (PPIs) and protein architectures with finer temporal resolution.

Exploration of quantitative site-specific serum O-glycoproteomics with isobaric labeling for the discovery of putative O-glycoprotein biomarkers.

Exploration study of site-specific isobaric-TMT-labeling quantitative serum O-glycoproteomics for the discovery of putative O-glycoprotein cancer biomarkers. Sera of 10 breast cancer patients was used as the exploration cohort. More abundant N-glycosylation was first removed with PNGase F. After tryptic digestion of de-N-glycosylated serum proteome, the TMT-labeled O-glycopeptides mixture was prepared and analyzed with RPLC-MS/MS. Site-specific qualitative and quantitative database search of O-glycopeptides was carried out with pGlyco 3.0. The same raw datasets were also searched with intact N-glycopeptide search engine GPSeeker to exclude possible interference of N-glycosylation. The final IDs were checked manually with GlcNAc-containing glycosite-determining fragment ions for confirmation. With the control of spectrum-level FDR≤1% and manual validation, 299 O-glycopeptides corresponding to 83 O-glycosites and 66 O-glycoproteins were identified, and 13 O-glycopeptides were found differentially expressed. Most interestingly, differential O-glycosylation was observed for IgG1 and IgG3, which is an interesting putative biomarker panel. Isobaric-labeling site-specific quantitative O-glycoproteomics is currently a state-of-the-art instrumental platform for discovery of putative seral cancer biomarkers. Differential seral O-glycosylation was observed in the IgG1 and IgG3.

A rare variant analysis framework using public genotype summary counts to prioritize disease-predisposition genes

Sequencing cases without matched healthy controls hinders prioritization of germline disease-predisposition genes. To circumvent this problem, genotype summary counts from public data sets can serve as controls. However, systematic inflation and false positives can arise if confounding factors are not controlled. We propose a framework, consistent summary counts based rare variant burden test (CoCoRV), to address these challenges. CoCoRV implements consistent variant quality control and filtering, ethnicity-stratified rare variant association test, accurate estimation of inflation factors, powerful FDR control, and detection of rare variant pairs in high linkage disequilibrium. When we applied CoCoRV to pediatric cancer cohorts, the top genes identified were cancer-predisposition genes. We also applied CoCoRV to identify disease-predisposition genes in adult brain tumors and amyotrophic lateral sclerosis. Given that potential confounding factors were well controlled after applying the framework, CoCoRV provides a cost-effective solution to prioritizing disease-risk genes enriched with rare pathogenic variants.

LinDA: linear models for differential abundance analysis of microbiome compositional data

Differential abundance analysis is at the core of statistical analysis of microbiome data. The compositional nature of microbiome sequencing data makes false positive control challenging. Here, we show that the compositional effects can be addressed by a simple, yet highly flexible and scalable, approach. The proposed method, LinDA, only requires fitting linear regression models on the centered log-ratio transformed data, and correcting the bias due to compositional effects. We show that LinDA enjoys asymptotic FDR control and can be extended to mixed-effect models for correlated microbiome data. Using simulations and real examples, we demonstrate the effectiveness of LinDA.

The breast tissue microbiome, stroma, immune cells and breast cancer.

Stromal and immune cell composition alterations in benign breast tissue associate with future cancer risk. Pilot data suggest the innate microbiome of normal breast tissue differs between women with and without breast cancer. Microbiome alterations might explain tissue microenvironment variations associated with disease status. Prospectively-collected sterile normal breast tissues from women with benign (n=16) or malignant (n=17) disease underwent 16SrRNA sequencing with Illumina MiSeq and Hybrid-denovo pipeline processing. Breast tissue was scored for fibrosis and fat percentages and immune cell infiltrates (lobulitis) classified as absent/mild/moderate/severe. Alpha and beta diversity were calculated on rarefied OTU data and associations analyzed with multiple linear regression and PERMANOVA. Breast tissue stromal fat% was lower and fibrosis% higher in benign disease versus cancer (median 30% versus 60%, p=0.01, 70% versus 30%, p=0.002, respectively). The microbiome varied with stromal composition. Alpha diversity (Chao1) correlated with fat% (r=0.38, p=0.02) and fibrosis% (r=-0.32, p=0.05) and associated with different microbial populations as indicated by beta diversity metrics (weighted UniFrac, p=0.08, fat%, p=0.07, fibrosis%). Permutation testing with FDR control revealed taxa differences for fat% in Firmicutes, Bacilli, Bacillales, Staphylococcaceae and genus Staphylococcus, and fibrosis% in Firmicutes, Spirochaetes, Bacilli, Bacillales, Spirochaetales, Proteobacteria RF32, Sphingomonadales, Staphylococcaceae, and genera Clostridium, Staphylococcus, Spirochaetes, Actinobacteria Adlercreutzia. Moderate/severe lobulitis was more common in cancer (73%) than benign disease (13%), p=0.003, but no significant microbial associations were seen. These data suggest a link between breast tissue stromal alterations and its microbiome, further supporting a connection between the breast tissue microenvironment and breast cancer.

Signatures of selection in Nelore cattle revealed by whole-genome sequencing data.

Nelore cattle breed was farmed worldwide due to its economic importance in the beef market and adaptation to the tropics. In Brazil, purebred Nelore animals (PO) receive a certificate from the breeders' association based on the animal's genealogy and morphological characterization. The top 20 to 30% of the superior animals are eligible to receive the Special Certificate of Identification and Production (CEIP), meaning animals from this category were selected and evaluated in a breeding program to improve economically important traits. We used whole-genome sequencing and approaches based on haplotype differentiation and allelic differentiation to detect regions of selection signatures in Nelore cattle by comparing animals from PO and CEIP categories. From a total of 150 animals, a hierarchical clustering analysis was performed to choose the more unrelated animals from each category (16 PO and 40 CEIP). The hapFLK statistic was performed, and extensions of hapFLK values were investigated considering continuous regions with significant q-values. The Weir and Cockerham's Fst estimator (wcFst) was computed using the GPAT++ software library. The total of 82,326 SNPs with hapFLK values passed the FDR control (q-value<0.05), and 718 segments were target as signatures of selection. A total of 1713 highly differentiated genomic regions were identified based on the segmentFst approach. The signatures of selection were spread across the genome. Annotation of overlapping selection signature regions between the two methods revealed 118 genes in common. A variant located within the 3' region of the BOLA-DRB3 gene was found as a promising candidate polymorphism. Within genomic regions that deserves attention, we found genes previously associated with adaptation to tropical environments (HELB), growth and navel size (HMGA2), fat deposition and domestication (IRAK3), and feed efficiency and postmortem carcass traits (GABRG3). The genes BOLA-DQA2, BOLA-DQB, BOLA-DQA5, BOLA-DQA1, BOLA-DRB3, ENSBTAG00000038397 on chromosome 23 are part of the Bovine Major Histocompatibility Complex (MHC) Class II gene family, representing good candidates for immune response and adaptation to tropical conditions. The BoLA family genes and the interaction of ROBO1 with SLIT genes appeared in the enrichment results. Genomic regions located in intronic regions were also identified and might play a regulatory role in traits under selection in PO and CEIP subpopulations. The regions here identified contribute to our knowledge regarding genes and variants that have an important role in complex traits selected in this breed.

Calibr improves spectral library search for spectrum-centric analysis of data independent acquisition proteomics

Identifying peptides and proteins from mass spectrometry (MS) data, spectral library searching has emerged as a complementary approach to the conventional database searching. However, for the spectrum-centric analysis of data-independent acquisition (DIA) data, spectral library searching has not been widely exploited because existing spectral library search tools are mainly designed and optimized for the analysis of data-dependent acquisition (DDA) data. We present Calibr, a spectral library search tool for spectrum-centric DIA data analysis. Calibr optimizes spectrum preprocessing for pseudo MS2 spectra, generating an 8.11% increase in spectrum–spectrum match (SSM) number and a 7.49% increase in peptide number over the traditional preprocessing approach. When searching against the DDA-based spectral library, Calibr improves SSM number by 17.6–26.65% and peptide number by 18.45–37.31% over two state-of-the-art tools on three different data sets. Searching against the public spectral library from MassIVE, Calibr improves state-of-the-art tools in SSM and peptide numbers by more than 31.49% and 25.24%, respectively, for two data sets. Our analyses indicate higher sensitivity of Calibr results from the use of various spectral similarity measures and statistical scores, coupled with machine learning-based statistical validation for FDR control. Calibr executable files including a graphical user-interface application are available at https://ms.iis.sinica.edu.tw/COmics/Software_CalibrWizard.html and https://sourceforge.net/projects/comics-calibr.

Multiattribute Glycan Identification and FDR Control for Glycoproteomics

Rapidly improving methods for glycoproteomics have enabled increasingly large-scale analyses of complex glycopeptide samples, but annotating the resulting mass spectrometry data with high confidence remains a major bottleneck. We recently introduced a fast and sensitive glycoproteomics search method in our MSFragger search engine, which reports glycopeptides as a combination of a peptide sequence and the mass of the attached glycan. In samples with complex glycosylation patterns, converting this mass to a specific glycan composition is not straightforward; however, as many glycans have similar or identical masses. Here, we have developed a new method for determining the glycan composition of N-linked glycopeptides fragmented by collisional or hybrid activation that uses multiple sources of information from the spectrum, including observed glycan B-type (oxonium) and Y-type ions and mass and precursor monoisotopic selection errors to discriminate between possible glycan candidates. Combined with false discovery rate estimation for the glycan assignment, we show that this method is capable of specifically and sensitively identifying glycans in complex glycopeptide analyses and effectively controls the rate of false glycan assignments. The new method has been incorporated into the PTM-Shepherd modification analysis tool to work directly with the MSFragger glyco search in the FragPipe graphical user interface, providing a complete computational pipeline for annotation of N-glycopeptide spectra with false discovery rate control of both peptide and glycan components that is both sensitive and robust against false identifications.