Perform tradeSeq tests in parallel

This function is a wrapper around tradeSeq test functions, such as associationTest or conditionTest, and allows multiple tests to be performed simultaneously, with all necessary parameters specified internally. It also organizes the output of each test into separate list elements, making it easy to access results. Finally, it calculates an adjusted p-value and orders based on decreasing Waldstat.

Usage

tradeSeqTests(
  models,
  tests = c("assoc", "pattern", "diffend", "early", "startvsend", "condition"),
  global = TRUE,
  pairwise = TRUE,
  lineages = TRUE,
  l2fc = 0,
  eigen.thresh = 0.01,
  n.points = 2 * nknots(models),
  knots = NULL,
  pseudotime.values = NULL,
  parallelized = FALSE,
  BPPARAM = NULL,
  tidy = TRUE,
  raw = FALSE,
  verbose = TRUE
)

Arguments

models

A SingleCellExperiment object containing the fitted GAM smoothers, computed using fitGAM, with or without conditions provided.

tests

Character. The names of one or several tests to perform. You may provide partial names, such as 'assoc' for associationTest, as long as a single match is found. The available tests are associationTest, patternTest, diffEndTest, earlyDETest, startVsEndTest and conditionTest. conditionTest is ignored if models was computed using fitGAM without conditions.

global

Logical. (from earlyDETest/startVsEndTest documentation) If TRUE, test for all pairwise comparisons/lineages simultaneously.

pairwise

Logical. (from earlyDETest documentation) If TRUE, test for all pairwise comparisons independently. Ignored in tests that do not have pairwise.

lineages

Logical. (from startVsEndTest documentation) If TRUE, test for all lineages independently. Ignored in tests that do not have lineages.

l2fc

Numeric. (from earlyDETest documentation) The log2 fold change threshold to test against. Note, that this will affect both the global test and the pairwise comparisons.

eigen.thresh

Numeric. (from earlyDETest documentation) Eigenvalue threshold for inverting the variance-covariance matrix of the coefficients to use for calculating the Wald test statistics. Lower values are more lenient to adding more information but also decrease computational stability. This argument should in general not be changed by the user but is provided for back-compatability. Set to 1e-8 to reproduce results of older version of 'tradeSeq'. Ignored in tests that do not have eigen.thresh.

n.points

Numeric. (from earlyDETest documentation) The number of points to be compared between lineages. Defaults to twice the number of knots. Ignored in tests that do not have n.points.

knots

Numeric or List. (from earlyDETest documentation) A vector of length 2 specifying the knots at the start and end of the region of interest. You may also provide a list of multiple elements (for example, list(c(2,4), c(3,4))) to repeat the test for multiple sets of knots. Ignored in tests that do not have knots.

pseudotime.values

Numeric or List. (from startVsEndTest documentation) A vector of length 2, specifying two pseudotime values to be compared against each other, for every lineage of the trajectory. @details Note that this test assumes that all lineages start at a pseudotime value of zero, which is the starting point against which the end point is compared. You may also provide a list of multiple elements (for example, list(c(8,12), c(6,14))) to repeat the test for multiple sets of pseudotime values. Ignored in tests that do not have pseudotime.values.

parallelized

Logical. If TRUE, the tests will be parallelized using BiocParallel. Please note that parallelization is complex and depends on your operating system (Windows users might not see a gain or might even experience a slowdown).

BPPARAM

A BiocParallelParam object to be used for parallelization. If NULL and parallelized = TRUE, the function will use a SerialParam object configured to use a single worker (core) and is therefore not parallelized, in order to prevent accidental use of large computation resources. Ignored if parallelized = FALSE.

tidy

Logical. If TRUE, a list is returned, with data.frame objects corresponding to each test divided into each global, pairwise and/or lineages comparison results. An adjusted p-value (False Discovery Rate) is also calculated and each data.frame object is ordered based on decreasing Waldstat.

raw

Logical. If TRUE, a list is returned, with data.frame objects corresponding to each test results.

verbose

Logical. If FALSE, does not print progress messages and output, but warnings and errors will still be printed.

Value

A list, with data.frame objects corresponding to each test results and/or data.frame objects corresponding to each test and each global, pairwise and/or lineages comparison results.