Knockoffs with Side Information

Ren, Zhimei; Candès, Emmanuel J.

doi:10.48550/arxiv.2001.07835

Cited by 8 publications

(11 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Fithian and Lei (2020) offers new technical tools for controlling FDR under dependence, including in settings with data-adaptive p-value weights, that may prove fruitful in relaxing the independence assumption. The adaptive knockoff method of Ren and Candès (2020) suggests another possible way forward by incorporating side information into multiple testing in supervised learning problems.…”

Section: Dependent P-valuesmentioning

confidence: 99%

AdaPT-GMM: Powerful and robust covariate-assisted multiple testing

Chao,

Fithian

2021

Preprint

View full text Add to dashboard Cite

We propose a new empirical Bayes method for covariate-assisted multiple testing with false discovery rate (FDR) control, where we model the local false discovery rate for each hypothesis as a function of both its covariates and p-value. Our method refines the adaptive p-value thresholding (AdaPT) procedure by generalizing its masking scheme to reduce the bias and variance of its false discovery proportion estimator, improving the power when the rejection set is small or some null p-values concentrate near 1. We also introduce a Gaussian mixture model for the conditional distribution of the test statistics given covariates, modeling the mixing proportions with a generic user-specified classifier, which we implement using a two-layer neural network. Like AdaPT, our method provably controls the FDR in finite samples even if the classifier or the Gaussian mixture model is misspecified. We show in extensive simulations and real data examples that our new method, which we call AdaPT-GMMg, consistently delivers high power relative to competing state-of-the-art methods. In particular, it performs well in scenarios where AdaPT is underpowered, and is especially well-suited for testing composite null hypothesis, such as whether the effect size exceeds a practical significance threshold.

show abstract

Section: Dependent P-valuesmentioning

confidence: 99%

AdaPT-GMM: Powerful and robust covariate-assisted multiple testing

Chao,

Fithian

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Another avenue of research would be the development of "wrapper" methods that enable the utilization of side-information by black-box supervised learning methods. For example, Ren and Candès [2020] develop methods for feature selection and ranking based on any feature importance statistic and any machine learning model can be enhanced by accounting for side-information.…”

Section: Discussionmentioning

confidence: 99%

“…This paradigm -flexible modeling plus calibration of a 1-dimensional parameter based on frequentist criteria-has proven to be fruitful for statistical applications going beyond the group-regularized ridge regression problem considered here. Further applications of this paradigm include feature selection in regression settings via knockoffs [Candès et al, 2018, Ren andCandès, 2020], multiple testing with side-information [Ignatiadis et al, 2016, Lei and Fithian, 2018, Ignatiadis and Huber, 2018, empirical Bayes shrinkage with side-information [Tan, 2016, Ignatiadis andWager, 2019] and conformal prediction [Vovk et al, 2005, Gupta et al, 2019.…”

Section: Discussionmentioning

confidence: 99%

$σ$-Ridge: group regularized ridge regression via empirical Bayes noise level cross-validation

Ignatiadis,

Lolas

2020

Preprint

View full text Add to dashboard Cite

Features in predictive models are not exchangeable, yet common supervised models treat them as such. Here we study ridge regression when the analyst can partition the features into K groups based on external side-information. For example, in high-throughput biology, features may represent gene expression, protein abundance or clinical data and so each feature group represents a distinct modality. The analyst's goal is to choose optimal regularization parameters λ = (λ1, . . . , λK ) -one for each group. In this work, we study the impact of λ on the predictive risk of group-regularized ridge regression by deriving limiting risk formulae under a high-dimensional random effects model with p n as n → ∞. Furthermore, we propose a data-driven method for choosing λ that attains the optimal asymptotic risk: The key idea is to interpret the residual noise variance σ 2 , as a regularization parameter to be chosen through cross-validation. An empirical Bayes construction maps the one-dimensional parameter σ to the K-dimensional vector of regularization parameters, i.e., σ → λ(σ). Beyond its theoretical optimality, the proposed method is practical and runs as fast as cross-validated ridge regression without feature groups (K = 1). 2 2 /2 [Hoerl and Kennard, 1970,

show abstract

“…To address this challenge, Ren and Candès (2020) developed an adaptive knockoff filter, extending the original knockoff filter of Barber and Candès (2015) and in order to increase its power by leveraging side information. This paper applies the method of Ren and Candès (2020) to analyze GWAS data from individuals with different ancestries , comparing its performance to that of a novel alternative approach. The main difference between the adaptive knockoff filter of Ren and Candès (2020) and the novel method proposed here in Section 2.3 is that the latter directly leverages side information while analyzing the raw genotype-phenotype data, while the former operates on pre-computed knockoff statistics.…”

Section: Related Workmentioning

confidence: 99%

Transfer learning in genome-wide association studies with knockoffs

Li¹,

Ren²,

Sabatti³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

This paper presents and compares alternative transfer learning methods that can increase the power of conditional testing via knockoffs by leveraging prior information in external data sets collected from different populations or measuring related outcomes. The relevance of this methodology is explored in particular within the context of genome-wide association studies, where it can be helpful to address the pressing need for principled ways to suitably account for, and efficiently learn from the genetic variation associated to diverse ancestries. Finally, we apply these methods to analyze several phenotypes in the UK Biobank data set, demonstrating that transfer learning helps knockoffs discover more numerous associations in the data collected from minority populations, potentially opening the way to the development of more accurate polygenic risk scores.

show abstract

Knockoffs with Side Information

Cited by 8 publications

References 47 publications

AdaPT-GMM: Powerful and robust covariate-assisted multiple testing

AdaPT-GMM: Powerful and robust covariate-assisted multiple testing

$σ$-Ridge: group regularized ridge regression via empirical Bayes noise level cross-validation

Transfer learning in genome-wide association studies with knockoffs

Contact Info

Product

Resources

About