filterGeneExpr {psichomics} | R Documentation |
Uses filterByExpr
to determine genes with sufficiently
large counts to retain for statistical analysis.
filterGeneExpr( geneExpr, minMean = 0, maxMean = Inf, minVar = 0, maxVar = Inf, minCounts = 10, minTotalCounts = 15 )
geneExpr |
Data frame or matrix: gene expression |
minMean |
Numeric: minimum of read count mean per gene |
maxMean |
Numeric: maximum of read count mean per gene |
minVar |
Numeric: minimum of read count variance per gene |
maxVar |
Numeric: maximum of read count variance per gene |
minCounts |
Numeric: minimum number of read counts per gene for a
worthwhile number of samples (check |
minTotalCounts |
Numeric: minimum total number of read counts per gene |
Boolean vector indicating which genes have sufficiently large counts
Other functions for gene expression pre-processing:
convertGeneIdentifiers()
,
normaliseGeneExpression()
,
plotGeneExprPerSample()
,
plotLibrarySize()
,
plotRowStats()
geneExpr <- readFile("ex_gene_expression.RDS") # Add some genes with low expression geneExpr <- rbind(geneExpr, lowReadGene1=c(rep(4:5, 10)), lowReadGene2=c(rep(5:1, 10)), lowReadGene3=c(rep(10:1, 10)), lowReadGene4=c(rep(7:8, 10))) # Filter out genes with low reads across samples geneExpr[filterGeneExpr(geneExpr), ]