RnaXSim: This function simulates an RNA-seq dataset based on a given...

In Linnorm: Linear model and normality based normalization and transformation method (Linnorm)

Description

This function simulates an RNA-seq dataset based on a given distribution.

Usage

RnaXSim(datamatrix, distribution = "NB", NumRep = 5, NumDiff = 2000,
  NumFea = 20000, showinfo = FALSE, DEGlog2FC = "Auto",
  MaxLibSizelog2FC = 0.5)

Arguments

datamatrix	Matrix. The matrix or data frame that contains your dataset. Each row is a feature (or Gene) and each column is a sample (or replicate). Raw Counts, CPM, RPKM, FPKM or TPM are supported. Undefined values such as NA are not supported. It is not compatible with log transformed datasets.This program assumes that all columns are replicates of the same sample.
distribution	Character: Defaults to "Poisson". This parameter controls the output distribution of the simulated RNA-seq dataset. It can be one of "Gamma" (Gamma distribution), "Poisson" (Poisson distribution), "LogNorm" (Log Normal distribution) or "NB" (Negative Binomial distribution).
NumRep	Integer: The number of replicates. This is half of the number of output samples. Defaults to 3.
NumDiff	Integer: The number of Differentially Changed Features. Defaults to 2000.
NumFea	Integer: The number of Total Features. Defaults to 20000.
showinfo	Logical: should we show data information on the console? Defaults to FALSE.
DEGlog2FC	"Auto" or Double: log 2 fold change threshold that defines differentially expressed genes. If set to "Auto," DEGlog2FC is defined at the level where ANOVA can get a q value of 0.05 with the average expression, where the data values are log1p transformed. Defaults to "Auto".
MaxLibSizelog2FC	Double: The maximum library size difference from the mean that is allowed, in terms of log 2 fold change. Set to 0 to prevent program from generating library size differences. Defaults to 0.5.

Value

This function returns a list that contains a matrix of count data in integer raw count and a vector that shows which genes are differentially expressed. In the matrix, each row is a gene and each column is a replicate. The first NumRep (see parameter) of the columns belong to sample 1, and the last NumRep (see parameter) of the columns belong to sample 2. There will be NumFea (see parameter) number of rows. The top NumCorr of genes will be positively or negatively correlated with each other (randomly); and they are evenly separated into groups. Each group is not intended to be correlated to each other, but, by chance, it can happen.

Examples

#Obtain example matrix:
data(SEQC)
expMatrix <- SEQC
#Example for Negative Binomial distribution
simulateddata <- RnaXSim(expMatrix, distribution="NB", NumRep=5, NumDiff = 200, NumFea = 2000)
#Example for Poisson distribution
simulateddata <- RnaXSim(expMatrix, distribution="Poisson", NumRep=5, NumDiff = 200, NumFea = 2000)
#Example for Log Normal distribution
simulateddata <- RnaXSim(expMatrix, distribution="LogNorm", NumRep=5, NumDiff = 200, NumFea = 2000)
#Example for Gamma distribution
simulateddata <- RnaXSim(expMatrix, distribution="Gamma", NumRep=5, NumDiff = 200, NumFea = 2000)

Linnorm documentation built on Nov. 8, 2020, 6:48 p.m.