forked from SomaLogic/SomaDataIO
-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- new function wrapper to allow users to merge in clinical variables to `soma_adat` objects easily - closes SomaLogic#80
- Loading branch information
Showing
6 changed files
with
227 additions
and
0 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,95 @@ | ||
#' Merge Clinical Data into SomaScan | ||
#' | ||
#' Occasionally, additional clinical data is obtained _after_ samples | ||
#' have been submitted to SomaLogic, or even after 'SomaScan' | ||
#' results have been delivered. | ||
#' This requires the new clinical variables, i.e. non-proteomic, data to be | ||
#' merged with 'SomaScan' data into a "new" ADAT prior to analysis. | ||
#' [merge_clin()] easily merges such clinical variables into an | ||
#' existing `soma_adat` object and is a simple wrapper around [dplyr::left_join()]. | ||
#' | ||
#' This funtionality also exists as a command-line tool (R script) contained | ||
#' in `merge_clin.R` that lives in the `cli/merge` system file directory. | ||
#' Please see: | ||
#' \itemize{ | ||
#' \item `dir(system.file("cli/merge", package = "SomaDataIO"), full.names = TRUE)` | ||
#' \item `vignette("cli-merge-tool", package = "SomaDataIO")` | ||
#' } | ||
#' | ||
#' @inheritParams params | ||
#' @param clin_data One of 2 options: | ||
#' \itemize{ | ||
#' \item A data frame containing clinical variables to merge into `x`, or | ||
#' \item A path to a file, typically a `*.csv`, | ||
#' containing clinical variables to merge into `x`. | ||
#' } | ||
#' @param by A character vector of variables to join by. | ||
#' See [dplyr::left_join()] for more details. | ||
#' @param by_class If `clin_data` is a file path, a named character vector | ||
#' of the variable its class. This ensures the `by-key` is compatible | ||
#' for the join. For example, `c(SampleId = "character")`. | ||
#' See [read.table()] for details about the `colClasses` argument, and | ||
#' the examples below. | ||
#' @param ... Additional parameters passed to [dplyr::left_join()]. | ||
#' @return An object of the same class as `x` with new clinical | ||
#' variables merged. | ||
#' @author Stu Field | ||
#' @seealso [dplyr::left_join()] | ||
#' @examples | ||
#' # retrieve clinical data | ||
#' clin_file <- system.file("cli/merge", "meta.csv", | ||
#' package = "SomaDataIO", | ||
#' mustWork = TRUE) | ||
#' clin_file | ||
#' | ||
#' # view clinical data to be merged: | ||
#' # 1) `group` | ||
#' # 2) `newvar` | ||
#' clin_df <- read.csv(clin_file, colClasses = c(SampleId = "character")) | ||
#' clin_df | ||
#' | ||
#' # create mini-adat | ||
#' apts <- withr::with_seed(123, sample(getAnalytes(example_data), 2L)) | ||
#' adat <- head(example_data, 9L) |> # 9 x 2 | ||
#' dplyr::select(SampleId, all_of(apts)) | ||
#' | ||
#' # merge clinical variables | ||
#' merged <- merge_clin(adat, clin_df, by = "SampleId") | ||
#' merged | ||
#' | ||
#' # Alternative syntax: | ||
#' # 1) pass file path | ||
#' # 2) merge on different variable names | ||
#' # 3) convert join type on-the-fly | ||
#' clin_file2 <- system.file("cli/merge", "meta2.csv", | ||
#' package = "SomaDataIO", | ||
#' mustWork = TRUE) | ||
#' | ||
#' id_type <- typeof(adat$SampleId) | ||
#' merged2 <- merge_clin(adat, clin_file2, # file path | ||
#' by = c(SampleId = "ClinKey"), # join on 2 variables | ||
#' by_class = c(ClinKey = id_type)) # match types | ||
#' merged2 | ||
#' @importFrom utils read.csv | ||
#' @importFrom dplyr left_join | ||
#' @export | ||
merge_clin <- function(x, clin_data, by = NULL, by_class = NULL, ...) { | ||
|
||
stopifnot("`x` must be a `soma_adat`." = is.soma_adat(x)) | ||
|
||
if ( inherits(clin_data, "data.frame") ) { | ||
clin_df <- clin_data | ||
} else if ( is.character(clin_data) && | ||
length(clin_data) == 1L && | ||
file.exists(clin_data) ) { | ||
clin_df <- normalizePath(clin_data, mustWork = TRUE) |> | ||
utils::read.csv(header = TRUE, colClasses = by_class, row.names = NULL, | ||
stringsAsFactors = FALSE) | ||
} else { | ||
stop( | ||
"Invalid `clin_data` argument: ", .value(class(clin_data)), | ||
"\n`clin_data` must be either a `data.frame` or file path.", call. = FALSE) | ||
} | ||
|
||
dplyr::left_join(x, clin_df, by = by, ...) | ||
} |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.
Oops, something went wrong.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,32 @@ | ||
|
||
# Setup ---- | ||
clin_file <- system.file("cli/merge", "meta.csv", package = "SomaDataIO", | ||
mustWork = TRUE) | ||
clin_df <- read.csv(clin_file, header = TRUE, colClasses = c(SampleId = "character")) | ||
apts <- withr::with_seed(123, sample(getAnalytes(example_data), 2L)) | ||
adat <- head(example_data, 9L) |> dplyr::select(SampleId, all_of(apts)) | ||
|
||
test_that("merge_clin() errors on bad `clin_data` argument", { | ||
merged <- merge_clin(adat, clin_df, by = "SampleId") | ||
expect_true(all(names(adat) %in% names(merged))) | ||
expect_equal(setdiff(names(merged), names(adat)), c("group", "newvar")) | ||
expect_equal(dim(merged), c(9, 5L)) | ||
expect_equal(sum(is.na(merged)), 8L) | ||
expect_equal(sum(merged$newvar, na.rm = TRUE), -1.779255) | ||
}) | ||
|
||
test_that("merge_clin() generates same result on `clin_data` argument", { | ||
expect_equal( | ||
merge_clin(adat, clin_df, by = "SampleId"), | ||
merge_clin(adat, clin_file, by = "SampleId", by_class = c(SampleId = "character")) | ||
) | ||
}) | ||
|
||
test_that("merge_clin() errors on bad `clin_data` argument", { | ||
expect_error( merge_clin(adat, letters) ) | ||
expect_error( merge_clin(adat, 1:10L) ) | ||
expect_error( merge_clin(adat, "Samples") ) | ||
expect_error( merge_clin(adat, NA) ) | ||
expect_error( merge_clin(adat, NA_character_) ) | ||
expect_error( merge_clin(data.frame(adat)) ) | ||
}) |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters