Rename to getOutlierIds, update stat articles

- renamed getFlaggedIds() to getOutlierIds()
- updated calls to preProcessAdat() in stat
  workflow articles
- removed recommendation to use pre-normalized
  data if scale factors correlate with biological
  endpoint

Author: scheidec

NAMESPACE

@@ -86,8 +86,8 @@ export(getAnalyteInfo)
 export(getAnalytes)
 export(getFeatureData)
 export(getFeatures)
-export(getFlaggedIds)
 export(getMeta)
+export(getOutlierIds)
 export(getSeqId)
 export(getSeqIdMatches)
 export(getSignalSpace)

R/0-declare-global-variables.R

@@ -29,6 +29,7 @@ utils::globalVariables(
     "seqid",
     "SubjectCount",
     "SubjectId",
+    "TargetFullName",
     "Type",
     "value",
     "Value"

R/getFlaggedIds.R renamed to R/getOutlierIds.R

@@ -24,9 +24,9 @@
 #' out_adat[12, apts] <- out_adat[12, apts] * 10
 #'
 #' om <- calcOutlierMap(out_adat)
-#' getFlaggedIds(om, out_adat, flags = 0.05, include = c("Sex", "Subarray"))
+#' getOutlierIds(om, out_adat, flags = 0.05, include = c("Sex", "Subarray"))
 #' @export
-getFlaggedIds <- function(x, flags = 0.05, data = NULL, include = NULL) {
+getOutlierIds <- function(x, flags = 0.05, data = NULL, include = NULL) {
 
   if ( !inherits(x, "outlier_map") ) {
     stop("Input `x` object must be class `outlier_map`!",

R/plotMap.R

@@ -101,7 +101,7 @@ plot.Map <- function(x, color.scheme = NULL,
   # determine color palette for the final figure
   color.scheme <- color.scheme %||% rev(grDevices::topo.colors(100))
   legend.ticks <- 2        # only 2 are required for boolean data
-  star_samples <- getFlaggedIds(obj, flags)$idx   # outlier samples to flag
+  star_samples <- getOutlierIds(obj, flags)$idx   # outlier samples to flag
 
   # define plot title & axis labels for starred samples
   title <- paste0(plot.info$title, "     ",

R/preProcessAdat.R

@@ -129,7 +129,7 @@ preProcessAdat <- function(adat,
 
   if( filter.outliers ) {
     # get count of sample level outliers by RFU
-    rfu_outliers <- suppressMessages(getFlaggedIds(calcOutlierMap(adat)))
+    rfu_outliers <- suppressMessages(getOutlierIds(calcOutlierMap(adat)))
     n_outliers <- nrow(rfu_outliers)
     adat <- adat |> filter(!dplyr::row_number() %in% rfu_outliers$idx)
 

inst/WORDLIST

@@ -99,10 +99,12 @@ frac
 funder
 ggplot
 gridlines
+homogenate
 intra
 leftrightarrow
 lifecycle
 lysate
+lysis
 magrittr
 medNormRef
 nd

man/calcOutlierMap.Rd

@@ -10,93 +10,93 @@ om <- calcOutlierMap(data)
 
 
 # Testing ------
-test_that("`getFlaggedIds()` returns error if `x` is not the required class", {
+test_that("`getOutlierIds()` returns error if `x` is not the required class", {
   expect_error(
-    getFlaggedIds(1:3),
+    getOutlierIds(1:3),
     "Input `x` object must be class `outlier_map`!"
   )
   expect_error(
-    getFlaggedIds("foo"),
+    getOutlierIds("foo"),
     "Input `x` object must be class `outlier_map`!"
   )
   expect_error(
-    getFlaggedIds(data.frame(x = 1)),
+    getOutlierIds(data.frame(x = 1)),
     "Input `x` object must be class `outlier_map`!"
   )
 })
 
-test_that("`getFlaggedIds()` returns error if `data` is not a df", {
+test_that("`getOutlierIds()` returns error if `data` is not a df", {
   expect_error(
-    getFlaggedIds(om, data = 1:3),
+    getOutlierIds(om, data = 1:3),
     "The `data` argument must be a `data.frame` object."
   )
 })
 
-test_that("`getFlaggedIds()` returns error if flags arg is not in [0, 1]", {
+test_that("`getOutlierIds()` returns error if flags arg is not in [0, 1]", {
   expect_error(
-    getFlaggedIds(om, flags = 1.1),
+    getOutlierIds(om, flags = 1.1),
     "`flags =` argument must be between 0 and 1!"
   )
   expect_error(
-    getFlaggedIds(om, flags = -0.1),
+    getOutlierIds(om, flags = -0.1),
     "`flags =` argument must be between 0 and 1!"
   )
 })
 
-test_that("`getFlaggedIds()` trips error if `include` not in `data`", {
+test_that("`getOutlierIds()` trips error if `include` not in `data`", {
   expect_error(
-    getFlaggedIds(om, 0.05, data, "foo"),
+    getOutlierIds(om, 0.05, data, "foo"),
     "All `include` must be in `data`."
   )
 })
 
-test_that("`getFlaggedIds()` returns 0 row df and msg if no obs are flagged", {
-  expect_message(outliers <- getFlaggedIds(om, 0.8),
+test_that("`getOutlierIds()` returns 0 row df and msg if no obs are flagged", {
+  expect_message(outliers <- getOutlierIds(om, 0.8),
                  "No observations were flagged at this flagging proportion:")
 
   expect_s3_class(outliers, "data.frame")
   expect_false(has_rn(outliers))
   expect_equal(outliers, data.frame(idx = numeric(0)))
 })
 
-test_that("`getFlaggedIds()` works on `calcOutlierMap` object, using default
+test_that("`getOutlierIds()` works on `calcOutlierMap` object, using default
           flags = 0.05, no included variables", {
-  outliers <- getFlaggedIds(om, data = data)
+  outliers <- getOutlierIds(om, data = data)
   expect_s3_class(outliers, "data.frame")
   expect_false(has_rn(outliers))
   expect_equal(outliers, data.frame(idx = c(12, 18)))
 })
 
-test_that("`getFlaggedIds()` works on `calcOutlierMap` object, using
+test_that("`getOutlierIds()` works on `calcOutlierMap` object, using
           flags = 0.02, one included variable", {
-  outliers <- getFlaggedIds(om, 0.02, data, "SampleId")
+  outliers <- getOutlierIds(om, 0.02, data, "SampleId")
   expect_s3_class(outliers, "data.frame")
   expect_false(has_rn(outliers))
   expect_equal(outliers, data.frame(idx      = c(12, 13, 18, 20),
                                     SampleId = c("14", "15", "21", "23")))
 })
 
-test_that("`getFlaggedIds()` works on `calcOutlierMap` object, using
+test_that("`getOutlierIds()` works on `calcOutlierMap` object, using
           flags = 0.1, multiple included variables", {
-  outliers <- getFlaggedIds(om, 0.1, data, c("SampleId", "Sex"))
+  outliers <- getOutlierIds(om, 0.1, data, c("SampleId", "Sex"))
   expect_s3_class(outliers, "data.frame")
   expect_false(has_rn(outliers))
   expect_equal(outliers, data.frame(idx         = 12,
                                     SampleId    = "14",
                                     Sex         = "M"))
 })
 
-test_that("`getFlaggedIds()` works on `calcOutlierMap` object with `data = NULL`", {
+test_that("`getOutlierIds()` works on `calcOutlierMap` object with `data = NULL`", {
   # data = NULL, include = NULL
-  outliers <- getFlaggedIds(om)
+  outliers <- getOutlierIds(om)
   expect_s3_class(outliers, "data.frame")
   expect_false(has_rn(outliers))
   expect_equal(outliers, data.frame(idx = c(12, 18)))
 })
 
-test_that("`getFlaggedIds()` `include` is ignored `data = NULL`", {
+test_that("`getOutlierIds()` `include` is ignored `data = NULL`", {
   expect_equal(
-    getFlaggedIds(om),
-    getFlaggedIds(om, data = NULL, include = "SampleId")  # test if `data = NULL`
+    getOutlierIds(om),
+    getOutlierIds(om, data = NULL, include = "SampleId")  # test if `data = NULL`
   )
 })

man/getFlaggedIds.Rd renamed to man/getOutlierIds.Rd

@@ -331,12 +331,12 @@ plot(om)
 The one sample in `filt_data` that we modified in the previous code chunk
 was identified as an outlier by RFU measurements with the `*`. 
 
-The `getFlaggedIds()` function is useful for extracting the
+The `getOutlierIds()` function is useful for extracting the
 sample IDs that were flagged as outliers in the plot. These sample IDs should be
 investigated further and possibly considered for removal prior to analysis.
 
 ```{r get-flagged-id}
-rfu_outliers <- getFlaggedIds(om)
+rfu_outliers <- getOutlierIds(om)
 rfu_outliers
 
 # drop one outlier sample
@@ -433,9 +433,8 @@ norm_vars |>
 ```
 
 If there *is* a considerable normalization bias observed in your ADAT,
-an alternate normalization or use of pre-normalized data should be used
-for univariate analyses, and any planned analyses may need to be
-modified to account for this bias.
+an alternate normalization should be used for univariate analyses, and any
+planned analyses may need to be modified to account for this bias.
 
 ------------------------------------------------------------------------
 

man/plot.Map.Rd

@@ -58,10 +58,10 @@ table(example_data$SampleType)
 cleanData <- example_data |>
   preProcessAdat(
     filter.features = TRUE,            # rm non-human protein features
-    filter.samples  = TRUE,            # rm control samples
-    data.qc         = "Sex",           # plot scale factors by covariate
+    filter.controls = TRUE,            # rm control samples
+    filter.qc       = TRUE,            # rm non-passing qc samples
     log.10          = TRUE,            # log10 transform
-    center.scale    = TRUE             # center, scale transform
+    center.scale    = TRUE             # center/scale analytes
   )
 
 # drop any missing values in Sex, and convert to binary 0/1 variable

man/preProcessAdat.Rd

@@ -58,10 +58,10 @@ table(example_data$SampleType)
 cleanData <- example_data |>
   preProcessAdat(
     filter.features = TRUE,            # rm non-human protein features
-    filter.samples  = TRUE,            # rm control samples
-    data.qc         = "Age",           # plot scale factors by covariate
+    filter.controls = TRUE,            # rm control samples
+    filter.qc       = TRUE,            # rm non-passing qc samples
     log.10          = TRUE,            # log10 transform
-    center.scale    = TRUE             # center, scale transform
+    center.scale    = TRUE             # center/scale analytes
   )
 
 # drop any missing Age values

tests/testthat/test-getFlaggedIds.R renamed to tests/testthat/test-getOutlierIds.R

@@ -60,10 +60,10 @@ table(example_data$SampleType)
 cleanData <- example_data |>
   preProcessAdat(
     filter.features = TRUE,            # rm non-human protein features
-    filter.samples  = TRUE,            # rm control samples
-    data.qc         = "Sex",           # plot scale factors by covariate
+    filter.controls = TRUE,            # rm control samples
+    filter.qc       = TRUE,            # rm non-passing qc samples
     log.10          = TRUE,            # log10 transform
-    center.scale    = TRUE             # center, scale transform
+    center.scale    = TRUE             # center/scale analytes
   )
 
 # drop any missing values in Sex

vignettes/articles/pre-processing.Rmd

@@ -58,10 +58,10 @@ table(example_data$SampleType)
 cleanData <- example_data |>
   preProcessAdat(
     filter.features = TRUE,            # rm non-human protein features
-    filter.samples  = TRUE,            # rm control samples
-    data.qc         = "Sex",           # plot scale factors by covariate
+    filter.controls = TRUE,            # rm control samples
+    filter.qc       = TRUE,            # rm non-passing qc samples
     log.10          = TRUE,            # log10 transform
-    center.scale    = TRUE             # center, scale transform
+    center.scale    = TRUE             # center/scale analytes
   )
 
 # drop any missing values in Sex