Spatial smoothing and hot spot detection for CGH data using the fused lasso.

By Robert Tibshirani, Pei Wang. Biostatistics, 2008

398 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/87b3519292c14b8b4c2ee9078145a85dbf310b1d

How researchers see this paper

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

In Genetics

In Molecular Biology

Citations in the corpus

A penalized matrix decomposition, with applications to sparse principal components and canonical correlation analysis. (2009, background): “Friedman et al. (2007), Tibshirani & Wang (2008) , and Hoefling (2009).”
The Generalized Lasso Problem and Uniqueness (2018, methodology): “The generalized lasso has been used as a modeling tool in numerous application areas, such as copy number variation analysis (Tibshirani and Wang, 2008), sMRI image classification (Xin et al.”
Online Learning for Matrix Factorization and Sparse Coding (2009, methodology): “This kind of regularization has proven to be useful for exploiting genomic data such as CGH arrays (Tibshirani and Wang, 2008).”
A Dynamic Programming Algorithm for the Fused Lasso and L0-Segmentation (2013, methodology): “This generalization, speciﬁcally the Poisson EFLSA, may be a useful extension of the work by Tibshirani and Wang (2008).”
Inference for single and multiple change‐points in time series (2013, neutral): “As commented by Tibshirani and Wang (2008), the LASSO can be applied even if p> n, and it has a unique solution assuming that no two predictors are collinear.”
A Direct Algorithm for 1-D Total Variation Denoising (2013, background): “Total variation denoising can be interpreted as pulling the antideriva-tive of the signal taut in a tube around it.”
PATHWISE COORDINATE OPTIMIZATION (2007, methodology): “Tibshirani and Wang (2007) report excellent results in the application of the fused lasso, finding that the method outperforms other popular techniques for this problem.”
Molecular Profiling of Breast Cancer Cell Lines Defines Relevant Tumor Models and Provides a Resource for Cancer Gene Discovery (2009, methodology): “DNA gains and losses were identified using the cghFLasso (R package for Fused Lasso) method [23], which controls the false discovery rate (FDR) by using normal-normal hybridization arrays to approximate the null distribution of the test statistics (see [23] for more details).”
Feature Selection for Classification: A Review (2014, methodology): “For example, in the study of arrayCGH [65, 66], the features (the DNA copy numbers along the genome) have the natural spatial order, and incorporating the structure information using an extension of the l1-norm outperforms the Lasso in both classification and feature selection.”
Exact post-selection inference for the generalized lasso path (2018, methodology): “The segment test contrasts were post-processed (i.e., “decluttered”) so as to exclude changepoints that occurred within 2 locations of each other—this only aﬀected the locations labeled E and F in the sparse 1d fused lasso model [40], with n = 990 points.”
SLEP: Sparse Learning with Efficient Projections (2011, background): “I t has found applications in comparative genomic hybridization [54, 65], prostate cancer analysis [64], ima ge denoising [12], and time-varying networks [1], where features can be ordered in some meaningful way.”
CNVkit: Genome-Wide Copy Number Detection and Visualization from Targeted DNA Sequencing (2016, methodology): “Alternatively, the HaarSeg algorithm [33] or Fused Lasso [34] can be used in place of CBS.”
Multiple Change-Point Estimation With a Total Variation Penalty (2010, methodology): “In terms of change-point estimation, it means, as already known; see, for example, Yao and Au (1989) or Lavielle and Moulines (2000), that we cannot have a perfect estimation of the change points.”
An efficient algorithm for a class of fused lasso problems (2010, methodology): “It has found applications in comparative genomic hybridization [25, 31], prostate cancer analysis [30], image denoising [8], and time-varying networks [1], where features can be ordered in”