Description Usage Arguments Details Value Author(s) References Examples
Accurately identifying differentially expressed (DE) genes from time course RNA-seq data has been of tremendous significance in creating a global picture of cellular function. DE genes from the time course RNA-seq data can be classified into two types, parallel DE genes (PDE) and non-parallel DE (NPDE) genes. The former are often biologically irrelevant, whereas the latter are often biologically interesting. In this package, we propose a negative binomial mixed-effects (NBME) model to identify both PDE and NPDE genes in time course RNA-seq data.
1 |
data.count |
a n by p matrix of expression values. Data should be appropriately normalized beforehand. |
group.label |
a vector indicating the experimental conditions of each time point. |
gene.names |
a vector containing all the gene names. |
exon.length |
a vector containing the length of exons, only used in exon level data. |
exon.level |
logical:indicating if this is an exon level dataset. Default is FALSE. |
pvalue |
logical:indicating if p-values are returned. Default is TRUE. |
Nonparallel differential expression(NPDE) genes and parallel differential expression(PDE) genes detection.
A list with components
sorted |
an object returned by timeSeq.sort function. It contains sorted Kullback Leibler Ratios(KLRs) or p-values for identifying DE genes. |
count |
the number of exons or replicates for each gene. |
NPDE |
the NPDE ratios or p-values. |
PDE |
the PDE ratios or p-values. |
genenames |
gene names. |
table |
gene expression values. |
data |
a n by p matrix of expression values. |
gene.names |
a vector including all the gene names. |
group.label |
a vector indicating the experimental conditions of each time point. |
group.length |
the total number of time points. |
group1.length |
the number of time points of condition one. |
group2.length |
the number of time points of condition two. |
exon.level |
logical:indicating if this is an exon level dataset. Default is FALSE. |
pvalue |
logical:indicating if p-values are returned. Default is TRUE. |
Fan Gao and Xiaoxiao Sun
Sun, Xiaoxiao, David Dalpiaz, Di Wu, Jun S. Liu, Wenxuan Zhong, and Ping Ma. "Statistical inference for time course RNA-Seq data using a negative binomial mixed-effect model." BMC Bioinformatics, 17(1):324, 2016.
Chong Gu. Model diagnostics for smoothing spline ANOVA models. Canadian Journal of Statistics, 32(4):347-358, 2004.
Chong Gu. Smoothing spline ANOVA models. Springer, second edition, 2013.
Chong Gu and Ping Ma. Optimal smoothing in nonparametric mixed-effect models. Annals of Statistics, 1357-1379, 2005.
Wood (2001) mgcv:GAMs and Generalized Ridge Regression for R. R News 1(2):20-25
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 | ####Data should be appropriately normalized beforehand####
##Exon level data (The p-values calculation is not supported)
data(pAbp)
attach(pAbp)
model.fit <- timeSeq(data.count,group.label,gene.names,exon.length,exon.level=TRUE,pvalue=FALSE)
#NPDE genes have large KLRs
model.fit$NPDE
detach(pAbp)
##Gene level data (three replicates)
data(simulate.dt)
attach(simulate.dt)
model.fit <- timeSeq(data.count,group.label,gene.names,exon.level=FALSE,pvalue=TRUE)
#p-values
model.fit$NPDE
|
There were 50 or more warnings (use warnings() to see the first 50)
[1] 0.8404626
[1] 4.056052e-102 1.601623e-216 8.095688e-86 3.998878e-112 8.449241e-278
[6] 9.973802e-01 7.780800e-01 9.026850e-01 1.000000e+00 9.278024e-01
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.