Article
Biochemical Research Methods
Javier E. Flores, Lisa M. Bramer, David J. Degnan, Vanessa L. Paurus, Yuri E. Corilo, Chaevien S. Clendinen
Summary: Gas chromatography-mass spectrometry (GC-MS) is an analytic method used to identify small molecules (e.g., metabolites) in metabolomics research. However, there is a risk of false identification or discovery that is not quantified. To address this, researchers propose a model-based framework for estimating the false discovery rate (FDR) among a set of identifications.
JOURNAL OF THE AMERICAN SOCIETY FOR MASS SPECTROMETRY
(2023)
Article
Biochemical Research Methods
Dominik Madej, Long Wu, Henry Lam
Summary: This study introduces a new method called CDD for FDR estimation, utilizing a fixed empirical null score distribution. Benchmarking CDD against decoy-based PeptideProphet showed similar accuracy and stability in retrieving correct PSMs. This finding highlights the potential of Big Data approaches for statistical analysis in proteomics and questions the necessity of dataset-specific target-decoy searches.
JOURNAL OF PROTEOME RESEARCH
(2022)
Article
Management
Sareh Nabi, Houssam Nassif, Joseph Hong, Hamed Mamani, Guido Imbens
Summary: Adding domain knowledge as a prior in learning systems has been shown to improve results. This study proposes a hierarchical empirical Bayes approach that addresses the challenges of lacking informative priors and controlling parameter learning rates. By learning empirical meta-priors and decoupling learning rates of different feature groups, the method improves performance and convergence time.
MANAGEMENT SCIENCE
(2022)
Article
Multidisciplinary Sciences
Ling Wang, Zongqiang Liao
Summary: In this study, a method for estimating normal mean problem with unknown sparsity and correlations in the signals is proposed. The method decomposes the covariance matrix of the observed signals into common dependence and weakly dependent error terms, reducing the correlations among the signals. Sparsity is estimated using an empirical Bayesian method based on the likelihood of the signals with the common dependence removed. The proposed algorithm outperforms the existing method in simulated examples and is consistent with findings from other studies when applied to Hapmap gene expressions data.
Article
Statistics & Probability
Nikolaos Ignatiadis, Stefan Wager
Summary: In this paper, the authors develop flexible and practical confidence intervals for empirical Bayes estimands. These intervals have asymptotic frequentist coverage, even when the estimands are only partially identified or when empirical Bayes point estimates converge very slowly.
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
(2022)
Article
Biochemical Research Methods
Sangjeong Lee, Heejin Park, Hyunwoo Kim
Summary: Accurately estimating the false discovery rate (FDR) is crucial in proteomics. The traditional target-decoy strategy (TDS) assumes the probabilities of spectra matching target or decoy peptides are identical, but in reality, they are not. Therefore, we propose cTDS, which estimates the FDR more accurately by considering the probability of spectra being incorrectly identified as target or decoy peptides.
BMC BIOINFORMATICS
(2022)
Article
Biochemical Research Methods
Arya Ebadi, Jack Freestone, William S. Noble, Uri Keich
Summary: Controlling the false discovery rate (FDR) in proteomics experiments using target decoy competition (TDC) only controls the average proportion of false discoveries. However, the actual proportion of false discoveries (FDP) may exceed the specified FDR threshold. We demonstrate this using real data and present two methods, FDP Stepdown and TDC Uniform Band, which help bridge the gap between controlling the expected FDR and the empirical FDP.
JOURNAL OF PROTEOME RESEARCH
(2023)
Article
Statistics & Probability
Etienne Roquain, Nicolas Verzelen
Summary: This paper presents theoretical foundations to understand the limitations of classical multiple testing theory in large scale experiments and proposes a procedure to properly learn the null distribution. The study focuses on Gaussian null distributions with unknown rescaling parameters and derives a procedure that mimics the performance of the idealized oracle. The results establish a phase transition at the sparsity boundary and provide insights for general location models.
ANNALS OF STATISTICS
(2022)
Article
Psychology, Multidisciplinary
Herbert Hoijtink
Summary: Researchers increasingly use Bayes factor for hypotheses evaluation, with NHBT being sensitive to the specification of the scale parameter of the prior distribution, while IHBT is not. Using recommended default values for scaling parameters in NHBT leads to unpredictable operation characteristics, but selecting the scaling parameter to bias the Bayes factor towards the null hypothesis by 19 if the observed effect size is zero can address this issue in some cases. However, this does not solve all problems associated with NHBT.
PSYCHOLOGICAL METHODS
(2022)
Article
Statistics & Probability
Rida Benhaddou, Matthew A. Connell
Summary: In this study, we approximate the classical Bayes estimator by truncating the generalized Laguerre series and estimate its coefficients by minimizing the prior risk of the estimator. We develop a strategy for selecting the parameter of the generalized Laguerre function basis to ensure our estimator has finite variance. We show that our generalized Laguerre empirical Bayes approach is asymptotically optimal in the minimax sense.
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS
(2023)
Review
Biochemical Research Methods
Mitra Ebrahimpoor, Jelle J. Goeman
Summary: Volcano plots are commonly used to select interesting discoveries, but they may lead to inflated false discovery rates. We demonstrate this issue with simulation experiments and data, and propose alternative approaches for multiple testing that do not inflate the false discovery rate.
BRIEFINGS IN BIOINFORMATICS
(2021)
Article
Statistics & Probability
Chang Liu, Yue Yang, Howard Bondell, Ryan Martin
Summary: In the context of high-dimensional linear regression models, this study introduces a new approach to address multicollinearity issues, typically achieving optimal posterior estimation and demonstrating superior performance compared to existing methods on real and simulated data.
Article
Biochemical Research Methods
Kyowon Jeong, Philipp T. Kaulich, Wonhyeuk Jung, Jihyung Kim, Andreas Tholey, Oliver Kohlbacher
Summary: Top-down proteomics provides more comprehensive proteoform-level information, but reliable data analysis remains challenging. The conventional FDR estimation method may not work at the proteoform level, and the precursor deconvolution error rate should be taken into account.
Article
Biochemical Research Methods
Yan Liu, Hao Liang, Quan Zou, Zengyou He
Summary: The identification of essential proteins is an important problem in bioinformatics. Existing methods have limitations in providing context-free and easily interpretable quantifications of centrality values, specifying proper thresholds, and controlling the quality of reported essential proteins. To overcome these limitations, this study formulates the essential protein discovery problem as a multiple hypothesis testing problem and presents a significance-based method named SigEP. Experimental results demonstrate that SigEP outperforms competing algorithms.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2022)
Article
Engineering, Mechanical
Enjian Cai, Yi Zhang
Summary: This paper proposes a novel perspective of phase estimation by utilizing image priors on phase patches. The method learns nonlocal self-similarity (NSS) prior from training images using the patch group based Gaussian Mixture Model (PG-GMM) learning algorithm, and optimizes the phase information using gaussian component selection and weighted sparse coding. The proposed method achieves high-quality magnifications and clearer time domain motion estimates.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)