Identify DE genes by method-specific thresholds

Applies specific filtering rules to each expression method result in the ExpressionResultSet object and returns the DE genes per tool.

Usage

expressionDefinition(
  resultTool,
  groups = c(""),
  lfcMinLimma = -2,
  lfcMaxLimma = 2,
  pValueLimma = 0.05,
  FLimma = 0.8,
  lfcMinSamseq = -2,
  lfcMaxSamseq = 2,
  qValueSamseq = 0.05,
  scoreDSamseq = 0.8,
  lfcMinDeseq2 = -2,
  lfcMaxDeseq2 = 2,
  pValueDeseq2 = 0.05,
  lfcMinEdger = -2,
  lfcMaxEdger = 2,
  pValueEdger = 0.05,
  probNoiseq = 0.8,
  lfcMaxKnowseq = 2,
  lfcMinKnowseq = -2,
  pValueKnowseq = 0.05,
  deClassEbseq = "DE",
  ppThresholdEbseq = 0.8,
  printResults = FALSE,
  pathOutput = "."
)

Arguments

resultTool

An ExpressionResultSet object.

groups

Character vector of group names or conditions.

lfcMinLimma

Minimum logFC threshold for limma.

lfcMaxLimma

Maximum logFC threshold for limma.

pValueLimma

P-value cutoff for limma.

FLimma

Minimum F-statistic for multi-group limma.

lfcMinSamseq

Minimum fold-change for SAMSeq.

lfcMaxSamseq

Maximum fold-change for SAMSeq.

qValueSamseq

q-value cutoff for SAMSeq.

scoreDSamseq

Score(d) threshold for SAMSeq multi-class.

lfcMinDeseq2

Minimum log2 fold-change for DESeq2.

lfcMaxDeseq2

Maximum log2 fold-change for DESeq2.

pValueDeseq2

P-value cutoff for DESeq2.

lfcMinEdger

Minimum logFC for edgeR.

lfcMaxEdger

Maximum logFC for edgeR.

pValueEdger

P-value cutoff for edgeR.

probNoiseq

Probability threshold for NOISeq.

lfcMaxKnowseq

Maximum logFC for KnowSeq.

lfcMinKnowseq

Minimum logFC for KnowSeq.

pValueKnowseq

P-value cutoff for KnowSeq.

deClassEbseq

DE class to filter for EBSeq (e.g. "DE" or "EE").

ppThresholdEbseq

Posterior probability threshold for EBSeq.

printResults

Logical. Whether to write results to disk.

pathOutput

Character. Directory path to save result tables.

Value

A list of data.frames with DE genes per method.

Examples

set.seed(42)
counts <- matrix(
  as.integer(c(
    rnbinom(200, mu = 10, size = 1),
    rnbinom(200, mu = 100, size = 1)
  )),
  nrow = 100,
  dimnames = list(
    paste0("gene", seq_len(100)),
    paste0("sample", seq_len(4))
  )
)
groups_info <- c("control", "control", "treated", "treated")
treats <- c("control", "treated")
res <- runExpression(
  numberReplics = 2,
  groupName = treats,
  rDataFrameCount = counts,
  controlDeseq2 = "control",
  contrastDeseq2 = "treated"
)
#> calcNormFactors has been renamed to normLibSizes
#> Using classic mode.
#> baySeq libsizes
#> baySeq getPriors
#> Finding priors...
#> Warning: The '@replicates' slot is not a factor; converting now.
#> done.
#> Finding posterior likelihoods...
#> Length of priorReps:0
#> Length of priorSubset:100
#> Length of subset:100
#> Length of postRows:100
#> Analysing part 1 of 1
#> Preparing data...
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> .
#> done.
#> Estimating likelihoods...
#> ...done!
#> .
#> done.
#> Available columns in results_raw:
#> Warning: ‘>=’ not meaningful for factors
#> calcNormFactors has been renamed to normLibSizes
#> Removing intercept from test coefficients
#> [1] "Computing (M,D) values..."
#> [1] "Computing probability of differential expression..."
#> Warning: `expect_is()` was deprecated in the 3rd edition.
#> ℹ Use `expect_type()`, `expect_s3_class()`, or `expect_s4_class()` instead
#> Warning: `expect_is()` was deprecated in the 3rd edition.
#> ℹ Use `expect_type()`, `expect_s3_class()`, or `expect_s4_class()` instead
#> Warning: `expect_is()` was deprecated in the 3rd edition.
#> ℹ Use `expect_type()`, `expect_s3_class()`, or `expect_s4_class()` instead
#> Initial number of DE patterns = 2
#> Final number of DE patterns = 2
#> converting counts to integer mode
#> estimating size factors
#> estimating dispersions
#> gene-wise dispersion estimates
#> mean-dispersion relationship
#> final dispersion estimates
#> fitting model and testing
#> Estimating sequencing depths...
#> Resampling to get new data matrices...
#> perm= 1
#> perm= 2
#> perm= 3
#> perm= 4
#> perm= 5
#> perm= 6
#> perm= 7
#> perm= 8
#> perm= 9
#> perm= 10
#> perm= 11
#> perm= 12
#> perm= 13
#> perm= 14
#> perm= 15
#> perm= 16
#> perm= 17
#> perm= 18
#> perm= 19
#> perm= 20
#> perm= 21
#> perm= 22
#> perm= 23
#> perm= 24
#> Number of thresholds chosen (all possible thresholds) = 11
#> Getting all the cutoffs for the thresholds...
#> Getting number of false positives in the permutation...
#> Differential expression analysis completed.
expDef <- expressionDefinition(res, treats)