'Gene shaving' as a method for identifying distinct sets of genes with similar expression patterns

By Trevor J. Hastie, Robert Tibshirani, Michael B. Eisen, Ash A. Alizadeh, R. Levy, Louis Staudt, W. Chan, David Botstein, Patrick O. Brown. Genome Biology, 2000

639 citations recorded by the source corpus.

BackgroundLarge gene expression studies, such as those conducted using DNA arrays, often provide millions of different pieces of data. To address the problem of analyzing such data, we describe a statistical method, which we have called 'gene shaving'. The method identifies subsets of genes with coherent expression patterns and large variation across conditions. Gene shaving differs from hierarchical clustering and other widely used methods for analyzing gene expression studies in that genes may belong to more than one cluster, and the clustering may be supervised by an outcome measure. The technique can be 'unsupervised', that is, the genes and samples are treated as unlabeled, or partially or fully supervised by using known properties of the genes or samples to assist in finding meaningful groupings.ResultsWe illustrate the use of the gene shaving method to analyze gene expression measurements made on samples from patients with diffuse large B-cell lymphoma. The method identifies a small cluster of genes whose expression is highly predictive of survival.ConclusionsThe gene shaving method is a potentially useful tool for exploration of gene expression data and identification of interesting clusters of genes worth further investigation.

Subfields: Genetics, Molecular Biology

View Paper: https://www.semanticscholar.org/paper/e22808d0205c2e763acfa3efd86a915050601e0b

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 mixture model-based approach to the clustering of microarray expression data (2002, background): “…validation, and understanding of various classes and subclasses of cancer; see, for example, Eisen et al. (1998), Ben-Dor et al. (1999) (2000), Alon et al. (1999), Golub et al. (1999), Hastie et al. (2000), Moler et al. (2000), Nguyen and Rocke (2001), and Xing and Karp (2001), among others.”
Computational cluster validation in post-genomic data analysis (2005, background): “Prominent examples include biclustering techniques [51] and gene shaving [33], which have both been specifically designed to deal with the particular challenges posed by gene expression data.”
Sparse principal component analysis via regularized low rank matrix approximation (2008, methodology): “(Hastie et al., 2000) and meta-genes (West, 2003).”
A high-fat diet coordinately downregulates genes required for mitochondrial oxidative phosphorylation in skeletal muscle. (2005, methodology): “After normalization, gene shaving (22), bootstrapping (23), and cluster analysis were performed (24), and the slide effect, dye effect, variety effect, and duplicate design were taken into account in an ANOVA model (25).”
Extracting Conserved Gene Expression Motifs from Gene Expression Data (2002, background): “Finally, Hastie et al. 17 present a technique called \gene shaving" that tries to extract coherent and small clusters of genes that vary as much as possible across the samples; in essence, their goals are complementary to ours.”
Prediction by Supervised Principal Components (2006, methodology): “Hastie et al. (2000) proposed “gene shaving” as a method for clustering genes.”
Defining transcription modules using large-scale gene expression data (2004, methodology): “results when applied to relatively small datasets, their utility for the analysis of large datasets appears to be limited by several drawbacks (Bittner et al., 1999; Cheng and Church, 2000; Gasch and Eisen, 2002; Getz et al., 2000; Hastie et al., 2000; Tanay et al., 2002).”
From patterns to pathways: gene expression data analysis comes of age (2002, background): “So-called 'fuzzy' clustering finds groups, but may allow elements to appear in more than one cluster or in no clusters at al”
Kernel Methods and Machine Learning (2014, background): “A “gene shaving” approach [93] in one which seeks subsets of genes with large variations across samples (or conditions) that have similar expression patterns by iteratively discarding the genes with the smallest variations was proposed.”
On Oblique Random Forests (2011, methodology): “We also compare both random forest scores with a t-test measuring class separation individually at every feature – similar to, for example, fMRI experiments, microarray, or tests on spectra [15].”
Finding large average submatrices in high dimensional data (2009, methodology): “The simplest, and most common, way of identifying associations in gene expression data is to independently cluster the rows and columns of the data matrix using a multivariate clustering procedure [Weinstein et al. (1997); Eisen et al. (1998); Tamayo et al. (1999); Hastie et al. (2000)].”
Sparse Principal Component Analysis (2006, methodology): “Hastie et al. (2000) proposed the so-called gene shaving techniques using PCA to cluster highly variable and coherent genes in microarray datasets.”
Singular value decomposition and principal component analysis (2003, methodology): “, 2001) and gene shaving (Hastie et al., 2000) are published SVD-based grouping algorithms; SVDMAN is free software available at http://home.”
Linear modes of gene expression determined by independent component analysis (2002, neutral): “Gene shaving (Hastie et al., 2000) determines optimal”
Generalized Thresholding of Large Covariance Matrices (2009, methodology): “Clustering genes using their correlations is a popular technique in gene expression data analysis (Eisen et al. 1998; Hastie et al. 2000).”