grouping_threshold added to the remove_noise function

RECETOX · Jul 9, 2024 · debf57c · debf57c
1 parent 10e8f9f
commit debf57c
Show file tree

Hide file tree

Showing 4 changed files with 36 additions and 4 deletions.
diff --git a/R/remove_noise.R b/R/remove_noise.R
@@ -60,6 +60,7 @@ load_data <- function(filename,
 #' @param intensity_weighted Whether to use intensity to weight mass density estimation.
 #' @param do.plot Indicates whether plot should be drawn.
 #' @param cache Whether to use cache
+#' @param grouping_threshold The maximum difference between two scans to be considered the same EIC. Default is Inf. 
 #' @return A matrix with four columns: m/z value, retention time, intensity, and group number.
 #' @export
 remove_noise <- function(filename,
@@ -71,7 +72,7 @@ remove_noise <- function(filename,
                      intensity_weighted,
                      do.plot,
                      cache,
-                     grouping_threshold = 0) {
+                     grouping_threshold = Inf) {
   raw.data <- load_file(filename)
 
   raw.prof <- adaptive.bin(
@@ -93,7 +94,7 @@ remove_noise <- function(filename,
 
   newprof <- newprof[newprof[, 4] %in% run.sel, ]
 
-  if (grouping_threshold > 0) {
+  if (grouping_threshold < Inf) {
     sorted_newprof <- newprof[order(newprof[,2]),]
     new_grps <- cumsum(c(0, diff(sorted_newprof[,2])) > grouping_threshold)
     sorted_newprof <- cbind(sorted_newprof, new_grps, deparse.level = 0)

diff --git a/conda/environment-dev.yaml b/conda/environment-dev.yaml
@@ -8,7 +8,7 @@ dependencies:
   - icu <=70.1
   - r-mass
   - r-rgl
-  - bioconductor-mzR ==2.28.0
+  - bioconductor-mzR ==2.36.0
   - r-splines2
   - r-doparallel
   - r-foreach

diff --git a/tests/remote-files/input.txt b/tests/remote-files/input.txt
@@ -6,4 +6,5 @@ https://gitlab.ics.muni.cz/umsa/umsa-files/-/raw/master/testdata/recetox-aplcms/
 https://gitlab.ics.muni.cz/umsa/umsa-files/-/raw/master/testdata/recetox-aplcms/input/RCX_08_shortened.mzML
 https://gitlab.ics.muni.cz/umsa/umsa-files/-/raw/master/testdata/recetox-aplcms/input/single_eic.mzml
 https://gitlab.ics.muni.cz/umsa/umsa-files/-/raw/master/testdata/recetox-aplcms/input/alg3.mzdata
-https://gitlab.ics.muni.cz/umsa/umsa-files/-/raw/master/testdata/recetox-aplcms/input/test_file.mzXML
+https://gitlab.ics.muni.cz/umsa/umsa-files/-/raw/master/testdata/recetox-aplcms/input/test_file.mzXML
+https://gitlab.ics.muni.cz/umsa/umsa-files/-/raw/master/testdata/recetox-aplcms/input/Tribrid_201106_009-QC1_1_NEG_FISABIO_single_eic.raw.mzML
diff --git a/tests/testthat/test-remove_noise.R b/tests/testthat/test-remove_noise.R
@@ -74,3 +74,33 @@ patrick::with_parameters_test_that(
     )
   )
 )
+
+test_that("remove noise works with grouping threshold", {
+  testdata <- file.path("..", "testdata")
+  input_path <- file.path(testdata,
+                          "input",
+                          "Tribrid_201106_009-QC1_1_NEG_FISABIO_single_eic.raw.mzML")
+
+  expected <- tibble(group_number = c(1, 2, 3, 5, 6, 7, 8, 9),
+                     n = c(67, 73, 3, 39, 2, 6, 3, 7))
+
+  sut <- remove_noise(
+    input_path,
+    min_pres = 0.8,
+    min_run = 0.2,
+    mz_tol = 5e-05,
+    baseline_correct = 0.0,
+    baseline_correct_noise_percentile = 0.05,
+    intensity_weighted = FALSE,
+    do.plot = FALSE,
+    cache = FALSE,
+    grouping_threshold = 4
+  )
+
+  actual <- sut %>%
+    mutate(group = factor(group_number)) %>%
+    group_by(group_number) %>%
+    summarize(n = n())
+
+  expect_equal(actual, expected)
+})