Multiple two-sample testing under arbitrary covariance dependency with an application in imaging mass spectrometry

Vladimir Vutov,Thorsten Dickhaus

DOI: https://doi.org/10.48550/arXiv.2108.08123

2021-08-18

Abstract:Large-scale hypothesis testing has become a ubiquitous problem in high-dimensional statistical inference, with broad applications in various scienfitic disciplines. One relevant application is constituted by imaging mass spectrometry (IMS) association studies, where a large number of tests are performed simultaneously in order to identify molecular masses that are associated with a particular phenotype, e. g., a cancer subtype. Mass spectra obtained from Matrix-assisted laser desorption/ionization (MALDI) experiments are dependent, when considered as statistical quantities. False discovery proportion (FDP) control under arbitrary dependency structure among test statistics is an active topic in modern multiple testing research. In this context, we are concerned with the evaluation of associations between the binary outcome variable (describing the phenotype) and multiple predictors derived from MALDI measurements. We propose an inference procedure in which the correlation matrix of the test statistics is utilized. The approach is based on multiple marginal models (MMM). Specifically, we fit a marginal logistic regression model for each predictor individually. Asymptotic joint normality of the stacked vector of the marginal regression coefficients is established under standard regularity assumptions, and their (limiting) correlation matrix is estimated. The proposed method extracts common factors from the resulting empirical correlation matrix. Finally, we estimate the realized FDP of a thresholding procedure for the marginal $p$-values. We demonstrate a practical application of the proposed workflow to MALDI IMS data in an oncological context.

Methodology

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to conduct multi - sample hypothesis testing under any covariance - dependence structure, especially its application in Imaging Mass Spectrometry (IMS). Specifically, researchers are concerned with controlling the False Discovery Proportion (FDP) in a large number of hypothesis tests. In particular, when analyzing Matrix - Assisted Laser Desorption/Ionization (MALDI) mass spectrometry data, how to effectively identify molecular masses related to specific phenotypes (such as a certain cancer subtype). The core problem of the paper is to propose a new statistical method. This method can use the correlation matrix of test statistics to improve the efficiency of multiple tests under the premise of considering the dependence between variables. This method is based on Multiple Marginal Models (MMM) and is achieved by fitting the marginal logistic regression model for each predictor variable. In addition, the paper also proposes a method for estimating the actual FDP and applies it to an actual case of MALDI - IMS data to verify the effectiveness of the method.

Multiple two-sample testing under arbitrary covariance dependency with an application in imaging mass spectrometry

Multiple multi-sample testing under arbitrary covariance dependency

Testing and Support Recovery of Multiple High-Dimensional Covariance Matrices with False Discovery Rate Control

Multiple Testing of Submatrices of a Precision Matrix with Applications to Identification of Between Pathway Interactions

A New Multiple Testing Method in the Dependent Case

Integrative High Dimensional Multiple Testing with Heterogeneity under Data Sharing Constraints

Large-Scale Simultaneous Testing Using Kernel Density Estimation

Projection-based two-sample inference for sparsely observed multivariate functional data

Large-Scale Multiple Testing of Correlations

Optimal False Discovery Rate Control for Large Scale Multiple Testing with Auxiliary Information

Simultaneous Inference Using Multiple Marginal Models

Large-Scale Simultaneous Testing of Cross-Covariance Matrices with Applications to PheWAS

Large-scale adaptive multiple testing for sequential data controlling false discovery and nondiscovery rates

Two-Sample Covariance Matrix Testing and Support Recovery in High-Dimensional and Sparse Settings

Testing a Large Number of Composite Null Hypotheses Using Conditionally Symmetric Multidimensional Gaussian Mixtures in Genome-Wide Studies

Two-Sample Tests for High-Dimensional Linear Regression with an Application to Detecting Interactions.

Application of multiple statistical tests to enhance mass spectrometry-based biomarker discovery

An exact projection pursuit-based algorithm for multivariate two-sample nonparametric testing applicable to retrospective and group sequential studies

Incorporation of Sparsity Information in Large-scale Multiple Two-sample $t$ Tests

Joint testing and false discovery rate control in high-dimensional multivariate regression.

Testing Independence with High-Dimensional Correlated Samples