RBDA: Rank Based Differential Expression Analysis This function...
In rramaker/RBDA: Rank Based Differential Expression Analysis

Description Usage Arguments Value Examples

View source: R/RBDA.R

Rank Based Differential Expression Analysis This function identifies differentially expressed genes from a raw count table

1 2	RBDA(countData, colData, testVariable, batch_family_variable = NULL, randomSeed = 1990, minP = 5e-06, numCores = 4)

`countData`	A raw count table (dataframe or matrix) with samples as columns and genes as rows.
`colData`	A dataframe with samples as rows and columns factor variables indicating experimental groups.
`testVariable`	Character indicating the column name in colData that reresents the variable of interest for differential expression analysis.
`batch_family_variable`	This paramater is used in N=1 cases only. Character that specifies a column name in colData that indicates other samples sequenced in the same batch as the sample of interest or related family members. It can be used as a downstream filter when comparing one sample to a large reference set. Defaults to NULL.
`randomSeed`	A numeric indicating a random seed for reproducible analysis. Defaults to 1990
`minP`	A numeric between 0 and one indicating the minimum possible P-value computed via random sampling from the countTable. The larger this number is the faster the compute time will be. Defaults to 0.000005.
`numCores`	A numeric indicating the number of cores used for computing. Most laptops can readily handle up to 4. Defaults to 4.

Returns a data frame containing differential expression results. The first column, "variance_rank," indicates the percentile rank of a gene's rank variance. Low numbers indicate a gene exhibits very low variance across samples. The second column "test_statistic" is the RBDA test statistic. The third column "p_value" indicates the probability of observing the test statistic by chance alone. A fourth column, "min_batch_fam_p_value," will only result if the batch_family_variable parameter is specified and will provide the minimum p-value computed for any samples indicated by this factor variable in colData.

1
2
3

data("BRCA_Counts")
colData<-data.frame(testVariable=as.factor(c(rep(0,5),rep(1,5))), row.names=colnames(BRCA_Counts))
RBDA(countData = BRCA_Counts,colData,testVariable = "testVariable", randomSeed = 1990, numCores=1)