Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions .Rbuildignore
Original file line number Diff line number Diff line change
Expand Up @@ -4,3 +4,7 @@
^README\.Rmd$
^\.github$
^codecov\.yml$
^pkgdown$
^README\.html$
^doc$
^Meta$
2 changes: 2 additions & 0 deletions .gitignore
Original file line number Diff line number Diff line change
Expand Up @@ -9,3 +9,5 @@ dev
inst/doc
vignettes/*.html
vignettes/*.R
/doc/
/Meta/
14 changes: 9 additions & 5 deletions DESCRIPTION
Original file line number Diff line number Diff line change
Expand Up @@ -15,17 +15,21 @@ BugReports: https://github.com/microbiome/daa/issues/
biocViews: Software, Microbiome
Encoding: UTF-8
Roxygen: list(markdown = TRUE)
RoxygenNote: 7.3.2
RoxygenNote: 7.3.3
Depends:
R (>= 4.5.0)
R (>= 4.5.0),
mia
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

mia could be dropped if possible to avoid dependencies.

Add SCE and SE as "Depends"

Imports:
dplyr
dplyr,
methods,
rstatix,
S4Vectors,
SingleCellExperiment,
SummarizedExperiment
Suggests:
BiocStyle,
knitr,
RefManageR,
rmarkdown,
sessioninfo,
testthat
Config/testthat/edition: 3
VignetteBuilder: knitr
23 changes: 23 additions & 0 deletions NAMESPACE
Original file line number Diff line number Diff line change
@@ -1,2 +1,25 @@
# Generated by roxygen2: do not edit by hand

export(addTtest)
export(addWilcoxon)
export(getTtest)
export(getWilcoxon)
exportMethods(addTtest)
exportMethods(addWilcoxon)
exportMethods(getTtest)
exportMethods(getWilcoxon)
importFrom(S4Vectors,DataFrame)
importFrom(SingleCellExperiment,altExp)
importFrom(SummarizedExperiment,assay)
importFrom(SummarizedExperiment,colData)
importFrom(SummarizedExperiment,rowData)
importFrom(dplyr,across)
importFrom(dplyr,all_of)
importFrom(dplyr,arrange)
importFrom(dplyr,group_by)
importFrom(methods,callNextMethod)
importFrom(mia,meltSE)
importFrom(rstatix,adjust_pvalue)
importFrom(rstatix,t_test)
importFrom(rstatix,wilcox_test)
importFrom(stats,as.formula)
25 changes: 25 additions & 0 deletions R/AllGenerics.R
Original file line number Diff line number Diff line change
@@ -0,0 +1,25 @@
# This file contains all the generics

#' @rdname getWilcoxon
#' @export
setGeneric("getWilcoxon", signature = "x", function(x, ...) {
standardGeneric("getWilcoxon")
})

#' @rdname getWilcoxon
#' @export
setGeneric("addWilcoxon", signature = "x", function(x, ...) {
standardGeneric("addWilcoxon")
})

#' @rdname getTtest
#' @export
setGeneric("getTtest", signature = "x", function(x, ...) {
standardGeneric("getTtest")
})

#' @rdname getTtest
#' @export
setGeneric("addTtest", signature = "x", function(x, ...) {
standardGeneric("addTtest")
})
156 changes: 156 additions & 0 deletions R/getTtest.R
Original file line number Diff line number Diff line change
@@ -0,0 +1,156 @@
#' @name
#' getTtest
#'
#' @title
#' Perform t-test
#'
#' @description
#' These functions perform t-test to compare values between two groups.
#'
#' @details
#' By default, Welch's t-test is used which does not assume equal variances.
#' Set \code{var.equal = TRUE} for Student's t-test.
#'
#' Specify exactly one of \code{assay.type}, \code{row.var}, or \code{col.var}
#' to define the values to test.
#'
#' @return
#' \code{getTtest} returns a \code{DataFrame} with test results.
#' \code{addTtest} returns \code{x} with results added to
#' \code{metadata(x)}.
#'
#' @param x A \code{SummarizedExperiment} object.
#'
#' @param assay.type \code{Character scalar} or \code{NULL}. Specifies assay to
#' test. Tests are run per feature. (Default: \code{NULL})
#'
#' @param row.var \code{Character scalar} or \code{NULL}. Specifies variable
#' from \code{rowData(x)} to test. (Default: \code{NULL})
#'
#' @param col.var \code{Character scalar} or \code{NULL}. Specifies variable
#' from \code{colData(x)} to test. (Default: \code{NULL})
#'
#' @param formula \code{formula}. A formula specifying the grouping variable,
#' e.g., \code{~ SampleType}. The RHS specifies the comparison groups.
#' For >2 levels, pairwise comparisons are performed.
#'
#' @param split.by \code{Character vector} or \code{NULL}. Columns to split by.
#' Tests are run separately for each combination. (Default: \code{NULL})
Comment on lines +37 to +38
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I guess this is more rarely used. This could be "hidden" parameter under .... Less options is better for usability.

#'
#' @param pair.by \code{Character scalar} or \code{NULL}. Column for pairing
#' samples in paired tests. (Default: \code{NULL})
#'
#' @param features \code{Character vector} or \code{NULL}. Specific features
#' to test when using \code{assay.type}. (Default: \code{NULL})
#'
#' @param var.equal \code{Logical scalar}. Assume equal variances?
#' (Default: \code{FALSE})
#'
#' @param p.adjust.method \code{Character scalar}. Method for p-value
#' adjustment. (Default: \code{"fdr"})
#'
#' @param name \code{Character scalar}. Column name prefix for results.
#' (Default: \code{"ttest"})
#'
#' @param ... Additional arguments passed to \code{rstatix::t_test}.
#'
#' @examples
#' data(GlobalPatterns, package = "mia")
#' tse <- GlobalPatterns
#' tse <- tse[1:50, tse$SampleType %in% c("Feces", "Skin")]
#' grp <- as.character(tse$SampleType)
#' assay_mat <- SummarizedExperiment::assay(tse, "counts")
#' keep <- apply(assay_mat, 1, function(v) {
#' a <- v[grp == "Feces"]
#' b <- v[grp == "Skin"]
#' !(stats::var(a, na.rm = TRUE) == 0 && stats::var(b, na.rm = TRUE) == 0)
#' })
#' tse <- tse[keep, ]
Comment on lines +63 to +68
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is too complex for an example

#'
#' # Test assay values (per feature)
#' res <- getTtest(tse, assay.type = "counts", formula = ~SampleType)
#'
#' # Test colData variable (e.g., alpha diversity)
#' tse$numeric_col <- as.numeric(seq_len(ncol(tse)))
#' res <- getTtest(tse, col.var = "numeric_col", formula = ~SampleType)
#'
#' @seealso
#' \code{\link[rstatix:t_test]{rstatix::t_test}},
#' \code{\link[daa:getWilcoxon]{getWilcoxon}}
#'
#' @export
NULL

#' @rdname getTtest
#' @export
#' @importFrom SingleCellExperiment altExp
#' @importFrom methods callNextMethod
setMethod("getTtest", "SingleCellExperiment", function(x, ...) {
x <- .check_and_get_altExp(x, ...)
res <- callNextMethod(x, ...)
return(res)
})

#' @rdname getTtest
#' @export
#' @importFrom SummarizedExperiment assay colData rowData
#' @importFrom rstatix t_test adjust_pvalue
#' @importFrom S4Vectors DataFrame
Comment on lines +97 to +98
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Add these importFrom above the function that is using the package --> easier to keep track of them

setMethod(
"getTtest", "SummarizedExperiment",
function(x, assay.type = NULL, row.var = NULL, col.var = NULL,
formula, split.by = NULL, pair.by = NULL, features = NULL,
var.equal = FALSE, p.adjust.method = "fdr", ...) {
############################# Input check ##############################
group <- .check_input(
Comment on lines +99 to +105
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Run:

styler::style_file(path = "inst/pages/machine_learning.qmd", transformers = styler::tidyverse_style(indent_by = 4L))

x, assay.type, row.var, col.var, formula,
split.by, pair.by, features
)
if (!.is_a_bool(var.equal)) {
stop("'var.equal' must be TRUE or FALSE.", call. = FALSE)
}
########################### Input check end ############################
# Get y variable name (the column to test)
y <- c(assay.type, row.var, col.var)
# Get data based on source
df <- .get_data(
x, assay.type, row.var, col.var, group,
split.by, pair.by, features
)
# Run tests
paired <- !is.null(pair.by)
res <- .run_ttest(
df, y, group, split.by, paired, var.equal,
p.adjust.method, features, ...
)
return(res)
}
)

#' @rdname getTtest
#' @export
setMethod(
"addTtest", "SummarizedExperiment",
function(x, name = "ttest", ...) {
if (!.is_non_empty_string(name)) {
stop("'name' must be a single character value.", call. = FALSE)
}
Comment on lines +135 to +137
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

You could use .check_input as it harmonizes the input test and messages

res <- getTtest(x, ...)
S4Vectors::metadata(x)[[name]] <- res
return(x)
}
)

################################################################################
# Internal function using .calc_daa engine
################################################################################

#' @importFrom rstatix t_test
.run_ttest <- function(df, y, group, split.by, paired, var.equal,
p.adjust.method, features = NULL, ...) {
.calc_daa(
df = df, y = y, group = group, split.by = split.by,
paired = paired, FUN = rstatix::t_test,
p.adjust.method = p.adjust.method, features = features, var.equal = var.equal, ...
)
}
Loading
Loading