Merge pull request #472 from Olink-Proteomics/develop_product_bridgin…

…g_downstream

Convert output of cross-product bridge to be used in downstream functions

OlinkAnalyze/R/olink_normalization.R

@@ -82,6 +82,9 @@
 #' subset normalization.
 #' @param reference_medians Dataset with columns "OlinkID" and "Reference_NPX".
 #' Required for reference median normalization.
+#' @param format Boolean that controls whether the normalized dataset will be
+#' formatted for input to downstream analysis. Only applicable for cross-product
+#' bridge normalization.
 #'
 #' @return Tibble or ArrowObject with the normalized dataset.
 #'
@@ -222,7 +225,8 @@
 #'   overlapping_samples_df1 = overlap_samples_product,
 #'   df1_project_nr = "proj_ht",
 #'   df2_project_nr = "proj_3k",
-#'   reference_project = "proj_ht"
+#'   reference_project = "proj_ht",
+#'   format = FALSE
 #' )
 #' }
 #'
@@ -233,7 +237,8 @@ olink_normalization <- function(df1,
                                 df1_project_nr = "P1",
                                 df2_project_nr = "P2",
                                 reference_project = "P1",
-                                reference_medians = NULL) {
+                                reference_medians = NULL,
+                                format = FALSE) {
 
   # check input ----
   lst_check <- olink_norm_input_check(
@@ -306,6 +311,14 @@ olink_normalization <- function(df1,
         not_ref_cols = lst_check$not_ref_cols
       )
 
+      if (format == TRUE) {
+        df_norm <- olink_normalization_product_format(bridged_df = df_norm,
+                                                      df1 = df1,
+                                                      df1_project_nr = df1_project_nr,
+                                                      df2 = df2,
+                                                      df2_project_nr = df2_project_nr)
+      }
+
     } else if (lst_check$norm_mode == olink_norm_modes$subset) {
       # subset normalization ----
 

OlinkAnalyze/R/olink_normalization_product.R

@@ -640,3 +640,145 @@ olink_normalization_qs <- function(lst_df,
 
   return(df_qq_norm)
 }
+
+
+
+#' Formatting the output of olink_normalization_product for seamless use with
+#' downstream OA functions.
+#'
+#' @author
+#' Danai G. Topouza
+#'
+#' @description
+#' Replaces the NPX values of the non-reference project by the Median Centered
+#' or QS Normalized NPX, according to the Bridging Recommendation. Edits the
+#' BridgingRecommendation column to indicate whether an assay is NotBridgeable,
+#' NotOverlapping, MedianCentering, or QuantileSmoothing bridged. Replaces
+#' OlinkID by the concatenation of the Explore HT and Explore 3072 OlinkIDs to
+#' record the OlinkIDs from both projects for bridgeable assays. Assays that are
+#' NotBridgeable or NotOverlapping retain their original non-reference OlinkIDs
+#' and NPX values. Replaces SampleID with the concatenation of SampleID and
+#' Project to make unique sample IDs for downstream analysis. Removes internal
+#' and external controls. Removes  MedianCenteredNPX, QSNormalizedNPX,
+#' OlinkID_E3072 columns.
+#'
+#' @param bridged_df A "tibble" of Olink data in long format resulting from the
+#' olink_normalization_product function.
+#' @param df1 First dataset to be used for normalization, pre-normalization.
+#' Must match df1 used in olink_normalization product bridging.
+#' @param df2 Second dataset to be used for normalization, pre-normalization.
+#' Must match df2 used in olink_normalization product bridging.
+#' @param df1_project_nr Project name of first dataset. Must match name used in
+#' olink_normalization product bridging.
+#' @param df2_project_nr Project name of second dataset. Must match name used in
+#' olink_normalization product bridging.
+#'
+#' @return A "tibble" of Olink data in long format containing both input
+#' datasets with the bridged NPX quantifications, with the above
+#' modifications.
+#'
+#' @examples
+#' \donttest{
+#' # Bridge samples
+#' bridge_samples <- intersect(
+#'   x = unique(OlinkAnalyze:::data_ht_small$SampleID),
+#'   y = unique(OlinkAnalyze:::data_3k_small$SampleID)
+#' ) |>
+#'   (\(x) x[!grepl("CONTROL", x)])()
+#'
+#' # Run olink_normalization_product
+#' npx_br_data <- olink_normalization(
+#' df1 = OlinkAnalyze:::data_ht_small,
+#' df2 = OlinkAnalyze:::data_3k_small,
+#' overlapping_samples_df1 = bridge_samples,
+#' df1_project_nr = "Explore HT",
+#' df2_project_nr = "Explore 3072",
+#' reference_project = "Explore HT")
+#'
+#' # Format output
+#' npx_br_data_format <- OlinkAnalyze:::olink_normalization_product_format(
+#' bridged_df = npx_br_data,
+#' df1 = OlinkAnalyze:::data_ht_small,
+#' df2 = OlinkAnalyze:::data_3k_small,
+#' df1_project_nr = "Explore HT",
+#' df2_project_nr = "Explore 3072")
+#'
+#' }
+
+olink_normalization_product_format <- function(bridged_df,
+                                               df1,
+                                               df1_project_nr,
+                                               df2,
+                                               df2_project_nr) {
+
+  # Extract data from NotBridgeable assays
+  df_not_bridgeable <- bridged_df |>
+    dplyr::mutate(SampleID = paste0(.data[["SampleID"]],
+                                    "_",
+                                    .data[["Project"]])) |>
+    dplyr::filter(.data[["SampleType"]] == "SAMPLE") |> # Remove controls
+    dplyr::filter(.data[["AssayType"]] == "assay") |>
+    dplyr::filter(.data[["BridgingRecommendation"]] == "NotBridgeable") |>
+    dplyr::mutate(OlinkID = case_when(
+      Panel == "Explore_HT" ~ OlinkID,
+      Panel != "Explore_HT" ~ OlinkID_E3072
+    )) |>
+    dplyr::select(!c(.data[["MedianCenteredNPX"]],
+                       .data[["QSNormalizedNPX"]],
+                       .data[["OlinkID_E3072"]]))# Remove extra columns
+
+
+  # Extract data from non-overlapping assays
+  df1_no_overlap <- df1 |>
+    dplyr::filter(.data[["SampleType"]] == "SAMPLE") |> # Remove controls
+    dplyr::filter(!(.data[["OlinkID"]] %in%
+               unlist(eHT_e3072_mapping |>
+                        dplyr::select(dplyr::starts_with("OlinkID_"))))) |>
+    dplyr::mutate(Project = df1_project_nr) |>
+    dplyr::mutate(SampleID =
+             paste0(.data[["SampleID"]],"_",df1_project_nr)) |>
+    dplyr::mutate(BridgingRecommendation = "NotOverlapping")
+
+  df2_no_overlap <- df2 |>
+    dplyr::filter(.data[["SampleType"]] == "SAMPLE") |> # Remove controls
+    dplyr::filter(!(.data[["OlinkID"]] %in%
+               unlist(eHT_e3072_mapping |>
+                        dplyr::select(dplyr::starts_with("OlinkID_"))))) |>
+    dplyr::mutate(Project = df2_project_nr) |>
+    dplyr::mutate(SampleID =
+             paste0(.data[["SampleID"]],"_",df2_project_nr)) |>
+    dplyr::mutate(BridgingRecommendation = "NotOverlapping")
+
+  ### Keep the data following BridgingRecommendation
+  df_format <- bridged_df |>
+    dplyr::filter(.data[["SampleType"]] == "SAMPLE") |> # Remove controls
+    dplyr::mutate(SampleID = paste0(.data[["SampleID"]],
+                                     "_",
+                                     .data[["Project"]])) |>
+    dplyr::filter(!.data[["BridgingRecommendation"]] == "NotBridgeable") |>
+   dplyr::mutate(NPX = case_when(
+      .data[["BridgingRecommendation"]] == "MedianCentering" ~
+        .data[["MedianCenteredNPX"]],
+      .data[["BridgingRecommendation"]] == "QuantileSmoothing" ~
+        .data[["QSNormalizedNPX"]],
+      .default = .data[["NPX"]])) |>
+    dplyr::filter(.data[["AssayType"]] == "assay") |>
+    dplyr::mutate(OlinkID = paste0(.data[["OlinkID"]],
+                                   "_",
+                                   .data[["OlinkID_E3072"]])) |>
+    dplyr::select(!c(.data[["MedianCenteredNPX"]],
+                     .data[["QSNormalizedNPX"]],
+                     .data[["OlinkID_E3072"]]))# Remove extra columns
+
+  df_full <- rbind(df_format,
+                   df_not_bridgeable,
+                   df1_no_overlap,
+                   df2_no_overlap)
+
+  # Sort by Project
+  df_full <- df_full |>
+    dplyr::arrange(.data[["Project"]], .data[["SampleID"]])
+
+  return(df_full)
+}
+

OlinkAnalyze/tests/testthat/helper-get_example_olink_data.R

@@ -1,8 +1,9 @@
 # load example datasets for Olink Explore products
 get_example_data <- function(filename) {
   ref_norm_res_file <- test_path("data", filename)
+  print(ref_norm_res_file)
   #check that file exists
-  expect_true(file.exists(ref_norm_res_file))
+  expect_true(file.exists(ref_norm_res_file),label = ref_norm_res_file)
   # read rds data
   readRDS(file = ref_norm_res_file)
 }