Genome-wide association analysis by lasso penalized logistic regression

By Tong Tong Wu, Y. Chen, Trevor J. Hastie, E. Sobel, Kenneth L. Lange. Bioinform., 2009

796 citations recorded by the source corpus.

Subfields: Gastroenterology, Genetics, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/d8f1ffeb7bcad215fcac25b0b0defcbc573d46e3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Gastroenterology

In Genetics

In Statistics and Probability

Citations in the corpus

Discovering causal interactions using Bayesian network scoring and information gain (2016, methodology): “Researchers applied standard statistical techniques including logistic regression [17,18], and regularized logistic regression [19,20].”
A variational Bayes algorithm for fast and accurate multiple locus genome-wide association analysis (2010, methodology): “We also demonstrate that V-Bay can outperform a recently proposed multiple locus methods that uses the least absolute shrinkage and selection operator (lasso) penalty [14], a theoretically well founded and widely accepted method for high dimensional model selection.”
Accounting for linkage disequilibrium in genome-wide association studies: A penalized regression method (2013, methodology): “Recently, this approach has been applied to GWAS for selecting important SNPs [19].”
Measurement error in Lasso: impact and likelihood bias correction (2012, methodology): “As with perfectly measured covariates (Zhao and Yu (2006)), the left-hand sides of A and B involve the model error ε, which needs to be bounded, and terms involving β0 S0 .”
Analysing biological pathways in genome-wide association studies (2010, methodology): “In fact, hierarchical models using LASSO have been successfully applied in simultaneous multivariate analyses of all GWA study SNP”
Multi-population GWA mapping via multi-task regularized regression (2010, methodology): “In this section, we assume that the population structure in the samples is known from prior knowledge or analysis, and make use of this information during the association analysis to detect both the population-specific and shared causal mutations.”
Identifying demand effects in a large network of product categories (2015, methodology): “The Lasso method is well established (Buhlmann and van de Geer, 2011; Chatterjee and Lahiri, 2011) and shows good performance in various applied fields (Wu et al., 2009; Fan et al., 2011).”
Prioritizing GWAS results: A review of statistical methods and recommendations for their application. (2010, methodology): “Much of their work is largely preserved in logistic regression.(80) To move from ordinary to logistic regression, one simply substitutes the negative loglikelihood for the loss function.”
Power and Predictive Accuracy of Polygenic Risk Scores (2013, background): “The normal distribution simplifies some calculations, but various heavy-tailed distributions have also been proposed for GWAS data [38,39] and would lead to improved prediction if such models held in truth.”
Strong rules for discarding predictors in lasso-type problems (2010, methodology): “Wu et al. (2009) study !1 penalized logistic regression and build a set D to discard based on the inner products between the outcome and each feature. As with the strong rules, their construction does not guarantee that the variables in D actually have zero coefficients in the solution, and so after”
Genomic selection using regularized linear regression models: ridge regression, lasso, elastic net and their extensions (2012, methodology): “In such cases, performance may be enhanced by using lasso type models that allow for interactions between predictors and correlated observations [8,25], nonparameteric or semiparametric regularized regression models [26], or other procedures able to efficiently handle large numbers of interactions,…”
Complementarity Formulations of ' 0 -norm Optimization Problems (2013, background): “In recent years, there has been an increased interest in solving optimization problems that minimize or restrict the number of nonzero elements of the solution vector [2, 3, 8, 10, 11, 12, 34, 37].”
Making the Most of Bag of Words: Sentence Regularization with Alternating Direction Method of Multipliers (2014, background): “In genomics, where it is common to have millions of features, Kim & Xing (2008) and Wu et al. (2009), among others, have used sparse models to obtain better performance for the genome wide association mapping problem.”
Epigenetic Predictor of Age (2011, methodology): “regression was performed using the ‘penalized’ package of R[14].”
The Kernel Interaction Trick: Fast Bayesian Discovery of Pairwise Interactions in High Dimensions (2019, methodology): “To avoid the combinatorial issues resulting from an L 0 penalty, recent works (Bien et al., 2013; Lim & Hastie, 2015) have instead used L 1 penalties to encourage parameter sparsity for interaction models; L 1 penalties have a long history in high-dimensional linear regression (Chen et al., 1998; Ca”
Regularization Paths for Generalized Linear Models via Coordinate Descent (2010, background): “Several other researchers have also re-discovered coordinate descent, many for solving the same problems we address in this paper—notably Shevade and Keerthi (2003), Krishnapuram and Hartemink (2005), Genkin et al. (2007) and Wu et al. (2009).”
Panning for Gold: Model-free Knockoffs for High-dimensional Controlled Variable Selection (2016, background): “…has had tremendous success establishing other useful inferential guarantees such as model selection consistency under high-dimensional asymptotics in both parametric (e.g., Lasso (Zhao and Yu, 2006; Candès and Plan, 2009)) and nonparametric (e.g., random forests (Wager and Athey, 2016)) settings.”