Small-sample estimation of negative binomial dispersion, with applications to SAGE data.

By Mark D. Robinson, Gordon K. Smyth. Biostatistics, 2007

1,128 citations recorded by the source corpus.

Subfields: Management Science and Operations Research, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/944006623073b2ea016b1040c64a74e857974bfc

How researchers see this paper

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

In Management Science and Operations Research

In Statistics and Probability

Citations in the corpus

Classes and continua of hippocampal CA1 inhibitory neurons revealed by single-cell transcriptomics (2017, positive): “All three methods were based on the observation that RNA counts within a homogeneous population can be approximated using a negative binomial distribution (See methods; Lu et al., 2005; Robinson and Smyth, 2008).”
Differential expression analysis for sequence count data (2010, methodology): “Robinson and Smyth [11] employed a technique they called quantile-adjusted conditional maximum likelihood to find an unbiased estimate for the raw SCV.”
Classes and continua of hippocampal CA 1 inhibitory neurons revealed by single ‐ cell transcriptomics (2017, methodology): “All three methods were based on the observation that RNA counts within a homogeneous population can be approximated using a negative binomial distribution (See methods; Lu et al., 2005; Robinson and Smyth, 2008).”
Applications of next-generation sequencing to the study of biological invasions (2015, methodology): “As a result, a variety of methods have been proposed based on negative binomial distributions (Robinson et al., 2010), common dispersal models (Robinson and Smyth, 2008), and Bayesian techniques (Hardcastle and Kelly, 2010).”
edgeR v4: powerful differential analysis of sequencing data with expanded functionality and improved support for small counts and larger datasets (2024, neutral): “Obtaining an unbiased estimator of the NB-dispersion is not straightforward because regular maximum likelihood estimation is biased leading to under-estimated dispersions [ 5 ].”
A comparison of statistical methods for detecting differentially expressed genes from RNA-seq data. (2012, positive): “Robinson and Smyth (2008) proposed to use a common dispersion for all genes to achieve a better estimate of the dispersion parameter.”
Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2 (2014, neutral): “This adjustment, first used in the context of dispersion estimation for SAGE data [47] and then for HTS data [3] in edgeR, corrects for the negative bias of dispersion estimates from using the maximum likelihood estimates (MLE) for the fitted values μ̂ij (analogous to Bessel’s correction in the usua”
Effects of library size variance, sparsity, and compositionality on the analysis of microbiome data (2015, methodology): “These may additionally be useful for microbial marker gene data 193 (Anders & Huber 2010; Anders et al. 2013; Law et al. 2014; Love MI 2014; McMurdie & 194 Holmes 2014; Paulson et al. 2013; Robinson et al. 2010; Robinson & Smyth 2008).”
TCC: an R package for comparing tag count data with robust normalization strategies (2013, methodology): “For example, the R package edgeR [8] uses a global scaling method, the trimmed mean of M values (TMM) method [15], in the data normalization step and an exact test for the negative binomial (NB) distribution [16] in the identification step.”
voom: precision weights unlock linear model analysis tools for RNA-seq read counts (2014, background): “This is the same context as assumed by a number of previous articles [18, 20, 13, 21, 34].”
Detecting differential usage of exons from RNA-seq data (2012, methodology): “Subsequently, Robinson and coworkers presented the edgeR method (Robinson and Smyth 2007, 2008; Robinson et al. 2010b), which introduced the use of the negative binomial distribution to RNA-seq analysis.”
Finding consistent patterns: A nonparametric approach for identifying differential expression in RNA-Seq data (2013, methodology): “These methods are parametric: they assume that (maybe after some simple transformation) each Nij is drawn from a particular distribution, such as Gaussian ([4, 12]), Poisson ([16, 7, 32, 11, 14]), negative binomial ([23, 24, 22, 11, 1]), et al ([2, 15]).”
baySeq: Empirical Bayesian methods for identifying differential expression in sequence count data (2010, methodology): “Robinson and Smyth derived an ‘ exact test ’ method based on the negative binomial distribution [8], and further developed this approach using a moderated test statistic sharing information across genomic locations to stabilize dispersion estimation in small samples [9].”
A comparison of methods for differential expression analysis of RNA-seq data (2013, methodology): “The first suggestion [12] was to assume that all genes had the same dispersion parameter.”
A scaling normalization method for differential expression analysis of RNA-seq data (2010, neutral): “[12], a Poisson regression [6] with library size and TMM normalization and a Poisson exact test [8] with library size and TMM normalization.”