scholarly journals Analytic Pearson residuals for normalization of single-cell RNA-seq UMI data

2020 ◽  
Author(s):  
Jan Lause ◽  
Philipp Berens ◽  
Dmitry Kobak
Keyword(s):  
Single Cell ◽  
Negative Binomial ◽  
Negative Binomial Regression ◽  
Downstream Processing ◽  
Negative Control ◽  
Parameter Estimates ◽  
Rna Seq ◽  
Pearson Residuals ◽  
Binomial Regression ◽  
Rank One

AbstractStandard preprocessing of single-cell RNA-seq UMI data includes normalization by sequencing depth to remove this technical variability, and nonlinear transformation to stabilize the variance across genes with different expression levels. Instead, two recent papers propose to use statistical count models for these tasks: Hafemeister and Satija (2019) recommend using Pearson residuals from negative binomial regression, while Townes et al. (2019) recommend fitting a generalized PCA model. Here, we investigate the connection between these approaches theoretically and empirically, and compare their effects on downstream processing. We show that the model of Hafemeister and Satija (2019) produces noisy parameter estimates because it is overspecified (which is why the original paper employs post-hoc regularization). When specified more parsimoniously, it has a simple analytic solution equivalent to the rank-one Poisson GLM-PCA of Townes et al. (2019). Further, our analysis indicates that per-gene overdispersion estimates in Hafemeister and Satija (2019) are biased, and that the data analyzed in that paper are in fact consistent with constant overdispersion parameter across genes. We then use negative control data without biological variability to estimate the technical overdispersion of UMI counts, and find that across several different experimental protocols, the data suggest very moderate overdispersion. Finally, we argue that analytic Pearson residuals (or, equivalently, rank-one GLM-PCA or negative binomial regression after regularization) strongly outperform standard preprocessing for identifying biologically variable genes, and capture more biologically meaningful variation when used for dimensionality reduction, compared to other methods.

scholarly journals Analytic Pearson residuals for normalization of single-cell RNA-seq UMI data

2021 ◽  
Vol 22 (1) ◽  
Author(s):  
Jan Lause ◽  
Philipp Berens ◽  
Dmitry Kobak
Keyword(s):  
Single Cell ◽  
Negative Binomial ◽  
Negative Binomial Regression ◽  
Ground Truth ◽  
Downstream Processing ◽  
Negative Control ◽  
Parameter Estimates ◽  
Rna Seq ◽  
Pearson Residuals ◽  
Rank One

Abstract Background Standard preprocessing of single-cell RNA-seq UMI data includes normalization by sequencing depth to remove this technical variability, and nonlinear transformation to stabilize the variance across genes with different expression levels. Instead, two recent papers propose to use statistical count models for these tasks: Hafemeister and Satija (Genome Biol 20:296, 2019) recommend using Pearson residuals from negative binomial regression, while Townes et al. (Genome Biol 20:295, 2019) recommend fitting a generalized PCA model. Here, we investigate the connection between these approaches theoretically and empirically, and compare their effects on downstream processing. Results We show that the model of Hafemeister and Satija produces noisy parameter estimates because it is overspecified, which is why the original paper employs post hoc smoothing. When specified more parsimoniously, it has a simple analytic solution equivalent to the rank-one Poisson GLM-PCA of Townes et al. Further, our analysis indicates that per-gene overdispersion estimates in Hafemeister and Satija are biased, and that the data are in fact consistent with the overdispersion parameter being independent of gene expression. We then use negative control data without biological variability to estimate the technical overdispersion of UMI counts, and find that across several different experimental protocols, the data are close to Poisson and suggest very moderate overdispersion. Finally, we perform a benchmark to compare the performance of Pearson residuals, variance-stabilizing transformations, and GLM-PCA on scRNA-seq datasets with known ground truth. Conclusions We demonstrate that analytic Pearson residuals strongly outperform other methods for identifying biologically variable genes, and capture more of the biologically meaningful variation when used for dimensionality reduction.

scholarly journals Normalization and variance stabilization of single-cell RNA-seq data using regularized negative binomial regression

2019 ◽  
Author(s):  
Christoph Hafemeister ◽  
Rahul Satija
Keyword(s):  
Single Cell ◽  
Negative Binomial ◽  
Negative Binomial Regression ◽  
Parameter Estimates ◽  
Rna Seq ◽  
Modeling Framework ◽  
Variance Stabilization ◽  
Pearson Residuals ◽  
Binomial Regression ◽  
Biological Heterogeneity

AbstractSingle-cell RNA-seq (scRNA-seq) data exhibits significant cell-to-cell variation due to technical factors, including the number of molecules detected in each cell, which can confound biological heterogeneity with technical effects. To address this, we present a modeling framework for the normalization and variance stabilization of molecular count data from scRNA-seq experiments. We propose that the Pearson residuals from ’regularized negative binomial regression’, where cellular sequencing depth is utilized as a covariate in a generalized linear model, successfully remove the influence of technical characteristics from downstream analyses while preserving biological heterogeneity. Importantly, we show that an unconstrained negative binomial model may overfit scRNA-seq data, and overcome this by pooling information across genes with similar abundances to obtain stable parameter estimates. Our procedure omits the need for heuristic steps including pseudocount addition or log-transformation, and improves common downstream analytical tasks such as variable gene selection, dimensional reduction, and differential expression. Our approach can be applied to any UMI-based scRNA-seq dataset and is freely available as part of the R packagesctransform, with a direct interface to our single-cell toolkitSeurat.

scholarly journals Normalization and variance stabilization of single-cell RNA-seq data using regularized negative binomial regression

2019 ◽  
Vol 20 (1) ◽  
Author(s):  
Christoph Hafemeister ◽  
Rahul Satija
Keyword(s):  
Single Cell ◽  
Negative Binomial ◽  
Negative Binomial Regression ◽  
Parameter Estimates ◽  
Rna Seq ◽  
Modeling Framework ◽  
Variance Stabilization ◽  
Pearson Residuals ◽  
Binomial Regression ◽  
Biological Heterogeneity

AbstractSingle-cell RNA-seq (scRNA-seq) data exhibits significant cell-to-cell variation due to technical factors, including the number of molecules detected in each cell, which can confound biological heterogeneity with technical effects. To address this, we present a modeling framework for the normalization and variance stabilization of molecular count data from scRNA-seq experiments. We propose that the Pearson residuals from “regularized negative binomial regression,” where cellular sequencing depth is utilized as a covariate in a generalized linear model, successfully remove the influence of technical characteristics from downstream analyses while preserving biological heterogeneity. Importantly, we show that an unconstrained negative binomial model may overfit scRNA-seq data, and overcome this by pooling information across genes with similar abundances to obtain stable parameter estimates. Our procedure omits the need for heuristic steps including pseudocount addition or log-transformation and improves common downstream analytical tasks such as variable gene selection, dimensional reduction, and differential expression. Our approach can be applied to any UMI-based scRNA-seq dataset and is freely available as part of the R package , with a direct interface to our single-cell toolkit .

scholarly journals Approaching precision public health by automated syndromic surveillance in communities

PLoS ONE ◽  
2021 ◽  
Vol 16 (8) ◽  
pp. e0254479
Author(s):  
Ta-Chien Chan ◽  
Jia-Hong Tang ◽  
Cheng-Yu Hsieh ◽  
Kevin J. Chen ◽  
Tsan-Hua Yu ◽  
...  
Keyword(s):  
Regression Model ◽  
Negative Binomial ◽  
Negative Binomial Regression ◽  
Health Department ◽  
Negative Binomial Regression Model ◽  
Negatively Associated ◽  
Pearson Residuals ◽  
Binomial Regression ◽  
Taipei City ◽  
Icd 10

Background Sentinel physician surveillance in communities has played an important role in detecting early signs of epidemics. The traditional approach is to let the primary care physician voluntarily and actively report diseases to the health department on a weekly basis. However, this is labor-intensive work, and the spatio-temporal resolution of the surveillance data is not precise at all. In this study, we built up a clinic-based enhanced sentinel surveillance system named “Sentinel plus” which was designed for sentinel clinics and community hospitals to monitor 23 kinds of syndromic groups in Taipei City, Taiwan. The definitions of those syndromic groups were based on ICD-10 diagnoses from physicians. Methods Daily ICD-10 counts of two syndromic groups including ILI and EV-like syndromes in Taipei City were extracted from Sentinel plus. A negative binomial regression model was used to couple with lag structure functions to examine the short-term association between ICD counts and meteorological variables. After fitting the negative binomial regression model, residuals were further rescaled to Pearson residuals. We then monitored these daily standardized Pearson residuals for any aberrations from July 2018 to October 2019. Results The results showed that daily average temperature was significantly negatively associated with numbers of ILI syndromes. The ozone and PM2.5 concentrations were significantly positively associated with ILI syndromes. In addition, daily minimum temperature, and the ozone and PM2.5 concentrations were significantly negatively associated with the EV-like syndromes. The aberrational signals detected from clinics for ILI and EV-like syndromes were earlier than the epidemic period based on outpatient surveillance defined by the Taiwan CDC. Conclusions This system not only provides warning signals to the local health department for managing the risks but also reminds medical practitioners to be vigilant toward susceptible patients. The near real-time surveillance can help decision makers evaluate their policy on a timely basis.

scholarly journals ComBat-seq: batch effect adjustment for RNA-seq count data

2020 ◽  
Vol 2 (3) ◽  
Author(s):  
Yuqing Zhang ◽  
Giovanni Parmigiani ◽  
W Evan Johnson
Keyword(s):  
Differential Expression ◽  
Count Data ◽  
Statistical Power ◽  
Negative Binomial ◽  
Genomic Data ◽  
Negative Binomial Regression ◽  
Negative Binomial Regression Model ◽  
Rna Seq ◽  
Batch Effects ◽  
Binomial Regression

Abstract The benefit of integrating batches of genomic data to increase statistical power is often hindered by batch effects, or unwanted variation in data caused by differences in technical factors across batches. It is therefore critical to effectively address batch effects in genomic data to overcome these challenges. Many existing methods for batch effects adjustment assume the data follow a continuous, bell-shaped Gaussian distribution. However in RNA-seq studies the data are typically skewed, over-dispersed counts, so this assumption is not appropriate and may lead to erroneous results. Negative binomial regression models have been used previously to better capture the properties of counts. We developed a batch correction method, ComBat-seq, using a negative binomial regression model that retains the integer nature of count data in RNA-seq studies, making the batch adjusted data compatible with common differential expression software packages that require integer counts. We show in realistic simulations that the ComBat-seq adjusted data results in better statistical power and control of false positives in differential expression compared to data adjusted by the other available methods. We further demonstrated in a real data example that ComBat-seq successfully removes batch effects and recovers the biological signal in the data.

scholarly journals Sample size calculations for the differential expression analysis of RNA-seq data using a negative binomial regression model

2019 ◽  
Vol 18 (1) ◽  
Author(s):  
Xiaohong Li ◽  
Dongfeng Wu ◽  
Nigel G.F. Cooper ◽  
Shesh N. Rai
Keyword(s):  
Regression Model ◽  
Sample Size ◽  
Negative Binomial ◽  
Negative Binomial Regression ◽  
Wald Test ◽  
Maximum Likelihood Estimates ◽  
Negative Binomial Regression Model ◽  
Rna Seq ◽  
Sample Sizes ◽  
Binomial Regression

Abstract High throughput RNA sequencing (RNA-seq) technology is increasingly used in disease-related biomarker studies. A negative binomial distribution has become the popular choice for modeling read counts of genes in RNA-seq data due to over-dispersed read counts. In this study, we propose two explicit sample size calculation methods for RNA-seq data using a negative binomial regression model. To derive these new sample size formulas, the common dispersion parameter and the size factor as an offset via a natural logarithm link function are incorporated. A two-sided Wald test statistic derived from the coefficient parameter is used for testing a single gene at a nominal significance level 0.05 and multiple genes at a false discovery rate 0.05. The variance for the Wald test is computed from the variance-covariance matrix with the parameters estimated from the maximum likelihood estimates under the unrestricted and constrained scenarios. The performance and a side-by-side comparison of our new formulas with three existing methods with a Wald test, a likelihood ratio test or an exact test are evaluated via simulation studies. Since other methods are much computationally extensive, we recommend our M1 method for quick and direct estimation of sample sizes in an experimental design. Finally, we illustrate sample sizes estimation using an existing breast cancer RNA-seq data.

scholarly journals ComBat-Seq: batch effect adjustment for RNA-Seq count data

2020 ◽  
Author(s):  
Yuqing Zhang ◽  
Giovanni Parmigiani ◽  
W. Evan Johnson
Keyword(s):  
Differential Expression ◽  
Count Data ◽  
Statistical Power ◽  
Negative Binomial ◽  
Genomic Data ◽  
Negative Binomial Regression ◽  
Batch Effect ◽  
Rna Seq ◽  
Batch Effects ◽  
Binomial Regression

AbstractThe benefit of integrating batches of genomic data to increase statistical power in differential expression is often hindered by batch effects, or unwanted variation in data caused by differences in technical factors across batches. It is therefore critical to effectively address batch effects in genomic data. Many existing methods for batch effect adjustment assume continuous, bell-shaped Gaussian distributions for data. However in RNA-Seq studies where data are skewed, over-dispersed counts, this assumption is not appropriate and may lead to erroneous results. Negative binomial regression models have been used to better capture the properties of counts. We developed a batch correction method, ComBat-Seq, using negative binomial regression. ComBat-Seq retains the integer nature of count data in RNA-Seq studies, making the batch adjusted data compatible with common differential expression software packages that require integer counts. We show in realistic simulations that the ComBat-Seq adjusted data result in better statistical power and control of false positives in differential expression, compared to data adjusted by the other available methods. We further demonstrated in a real data example where ComBat-Seq successfully removes batch effects and recovers the biological signal in the data.

A control chart based on Pearson residuals for a negative binomial regression: application to infant mortality data

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Sandra García-Bustos ◽  
Nadia Cárdenas-Escobar ◽  
Ana Debón ◽  
César Pincay
Keyword(s):  
Infant Mortality ◽  
Control Chart ◽  
Negative Binomial ◽  
Negative Binomial Regression ◽  
Mortality Data ◽  
Negative Binomial Regression Model ◽  
Content Type ◽  
Pearson Residuals ◽  
Binomial Regression ◽  
The Mean

PurposeThe study aims to design a control chart based on an exponentially weighted moving average (EWMA) chart of Pearson's residuals of a model of negative binomial regression in order to detect possible anomalies in mortality data.Design/methodology/approachIn order to evaluate the performance of the proposed chart, the authors have considered official historical records of death of children of Ecuador. A negative binomial regression model was fitted to the data, and a chart of the Pearson residuals was designed. The parameters of the chart were obtained by simulation, as well as the performances of the charts related to changes in the mean of death.FindingsWhen the chart was plotted, outliers were detected in the deaths of children in the years 1990–1995, 2001–2006, 2013–2015, which could show that there are underreporting or an excessive growth in mortality. In the analysis of performances, the value of λ = 0.05 presented the fastest detection of changes in the mean death.Originality/valueThe proposed charts present better performances in relation to EWMA charts for deviance residuals, with a remarkable advantage of the Pearson residuals, which are much easier to interpret and calculate. Finally, the authors would like to point out that although this paper only applies control charts to Ecuadorian infant mortality, the methodology can be used to calculate mortality in any geographical area or to detect outbreaks of infectious diseases.

Sign in / Sign up

Export Citation Format

Share Document