Merge pull request #243 from MoTrPAC/enhance/write-metabolomics

Enhance/write metabolomics
MoTrPAC · Sep 23, 2024 · db20073 · db20073
2 parents 37bab43 + 857bc1a
commit db20073
Show file tree

Hide file tree

Showing 5 changed files with 59 additions and 36 deletions.
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -1,8 +1,8 @@
 Package: MotrpacBicQC
 Type: Package
 Title: QC/QA functions for the MoTrPAC community
-Version: 0.9.5
-Date: 2024-05-22
+Version: 0.9.51
+Date: 2024-08-07
 Author: MoTrPAC Bioinformatics Center
 Maintainer: David Jimenez-Morales <davidjm@stanford.edu>
 Description: R Package for the analysis of MoTrPAC datasets. 

diff --git a/R/metabolomics_qc.R b/R/metabolomics_qc.R
@@ -181,7 +181,7 @@ check_metadata_samples <- function(df,
   # filter only expected columns
   df <- filter_required_columns(df = df,
                                 type = "m_s",
-                                verbose = TRUE)
+                                verbose = verbose)
 
   # Check every column
   # sample_id: si
@@ -276,12 +276,17 @@ check_metadata_samples <- function(df,
   }
 
   if("acquisition_date" %in% colnames(df)){
-    if( any(grepl(":", df$acquisition_date)) ){
-      if(verbose) message("  + (i) Assuming `acquisition_date` is in `MM/DD/YYYY HH:MM:SS AM/PM` format. Validating:")
-      icdt <- validate_dates_times(df = df, column_name = "acquisition_date", verbose = verbose)
+    if(any(is.na(df$acquisition_date))){
+      if(verbose) message("   - (-) `acquisition_date` has NA values: FAIL")
+      ic <- ic + 1
     }else{
-      icdate <- validate_yyyymmdd_dates(df = df, date_column = "acquisition_date", verbose = verbose)
-      ic <- ic + icdate
+      if( any(grepl(":", df$acquisition_date)) ){
+        if(verbose) message("  + (i) Assuming `acquisition_date` is in `MM/DD/YYYY HH:MM:SS AM/PM` format. Validating:")
+        icdt <- validate_dates_times(df = df, column_name = "acquisition_date", verbose = verbose)
+      }else{
+        icdate <- validate_yyyymmdd_dates(df = df, date_column = "acquisition_date", verbose = verbose)
+        ic <- ic + icdate
+      }
     }
   }else{
     if(verbose) message("   - (-) `acquisition_date` column missed: FAIL")
@@ -1200,15 +1205,17 @@ load_metabolomics_batch <- function(input_results_folder,
   assay <- validate_assay(input_results_folder)
   tissue_code <- validate_tissue(input_results_folder)
 
-  total_issues <- validate_metabolomics(input_results_folder = input_results_folder, 
-                                        cas = cas, 
-                                        return_n_issues = TRUE, 
-                                        full_report = FALSE, 
-                                        f_proof = FALSE, 
-                                        verbose = FALSE)
+  total_issues <- 
+    validate_metabolomics(
+      input_results_folder = input_results_folder, 
+      cas = cas, 
+      return_n_issues = TRUE, 
+      full_report = FALSE, 
+      f_proof = FALSE, 
+      verbose = FALSE)
 
   if(total_issues > 0){
-    message("\n\tWARNING!!! Too many issues identified (", total_issues,"). This batch should not be processed until the issues are solved")
+    message("\tWARNING!!! Too many issues identified (", total_issues,"). This batch should not be processed until the issues are solved")
   }
 
   # Load Metabolomics----
@@ -1376,8 +1383,10 @@ combine_metabolomics_batch <- function(input_results_folder,
                                        verbose = TRUE){
 
   # Load all datasets
-  metab_dfs <- load_metabolomics_batch(input_results_folder = input_results_folder,
-                                       cas = cas)
+  metab_dfs <- 
+    load_metabolomics_batch(
+      input_results_folder = input_results_folder,
+      cas = cas, verbose = verbose)
 
   if(verbose) message("\n## MERGE")
   if(verbose) message("\nAll metabolomics datasets + basic phenotypic information")
@@ -1677,9 +1686,11 @@ write_metabolomics_releases <- function(input_results_folder,
   if(verbose) message("+ Writing out ", cas, " ", phase_details, " ", tissue_code, " ", assay, " files", appendLF = FALSE)
 
   # Load all datasets----
-  metab_dfs <- load_metabolomics_batch(input_results_folder = input_results_folder,
-                                       cas = cas,
-                                       verbose = FALSE)
+  metab_dfs <- 
+    load_metabolomics_batch(
+      input_results_folder = input_results_folder,
+      cas = cas,
+      verbose = FALSE)
 
   # Create output folder-------
   if (is.null(folder_root)){

diff --git a/R/misc.R b/R/misc.R
@@ -310,27 +310,37 @@ filter_required_columns <- function(df,
     colnames(df) <- tolower(colnames(df))
     missing_cols <- setdiff(emeta_metabo_coln_named, colnames(df))
     if (length(missing_cols) > 0) {
-      if(verbose) message("   - (-) `metadata_metabolite`: Expected COLUMN NAMES are missed: FAIL")
-      message(paste0("\t The following required columns are not present: `", paste(missing_cols, collapse = ", "), "`"))
+      if (verbose) message("   - (-) `metadata_metabolite`: Expected COLUMN NAMES are missed: FAIL")
+      message(paste0("\t The following required columns are not present: `", 
+                     paste(missing_cols, collapse = ", "), "`"))
     } else {
-      if(verbose) message("  + (+) All required columns present")
+      if (verbose) message("  + (+) All required columns present")
       df <- subset(df, select = emeta_metabo_coln_named)
     }
     return(df)
-
-  } else if (type == "m_s"){
+  } else if (type == "m_s") {
     emeta_sample_coln <- c("sample_id", "sample_type", "sample_order", "raw_file", "extraction_date", "acquisition_date", "lc_column_id")
+    required_cols <- setdiff(emeta_sample_coln, c("extraction_date", "acquisition_date", "lc_column_id"))
     missing_cols <- setdiff(emeta_sample_coln, colnames(df))
-
-    if (length(missing_cols) > 0) {
-      if(verbose) message("   - (-) `metadata_sample`: Expected COLUMN NAMES are missed: FAIL")
-      message(paste0("\t The following required columns are not present: `", paste(missing_cols, collapse = ", "), "`"))
+    missing_required_cols <- setdiff(required_cols, colnames(df))
+
+    if (length(missing_required_cols) > 0) {
+      if (verbose) message("   - (-) `metadata_sample`: Expected COLUMN NAMES are missed: FAIL")
+      message(paste0("\t The following required columns are not present: `", 
+                     paste(missing_required_cols, collapse = ", "), "`"))
     } else {
-      if(verbose) message("  + (+) All required columns present")
+      if (length(missing_cols) > 0) {
+        message("   - (-) `metadata_sample`: recently required COLUMN NAMES are missed: Adding with NA values: FAIL")
+        for (col in c("extraction_date", "acquisition_date", "lc_column_id")) {
+          if (!(col %in% colnames(df))) {
+            df[[col]] <- NA
+          }
+        }
+      }
+      if (verbose) message("  + (+) All required columns present")
       df <- subset(df, select = emeta_sample_coln)
     }
     return(df)
-
   } else if (type == "v_m"){
     emeta_sample_coln <- c("vial_label", "tmt_plex")
     if( all(emeta_sample_coln %in% colnames(df)) ){
@@ -348,10 +358,10 @@ filter_required_columns <- function(df,
         if(verbose) message("  + (+) All required columns present (tmt18 experiment)")
         df <- subset(df, select = emeta_sample_coln)
       }else{
-        if(verbose) message("   - (-) Expected COLUMN NAMES are missed: FAIL")
+        message("   - (-) Expected COLUMN NAMES are missed: FAIL")
       }
     }else{
-      if(verbose) message("   - (-) Expected COLUMN NAMES are missed: FAIL")
+      message("   - (-) Expected COLUMN NAMES are missed: FAIL")
     }
     return(df)
   } else if (type == "olproteins"){

diff --git a/R/validations.R b/R/validations.R
@@ -396,7 +396,9 @@ validate_dates_times <- function(df, column_name, verbose = TRUE) {
 #' validate_lc_column_id(df, column_name = "lc_column_id")
 #' 
 #' @export
-validate_lc_column_id <- function(df, column_name, verbose = TRUE) {
+validate_lc_column_id <- function(df, 
+                                  column_name, 
+                                  verbose = TRUE) {
 
   # issue counter
   ic <- 0

diff --git a/tests/testthat/test-metabolomics_qc.R b/tests/testthat/test-metabolomics_qc.R
@@ -7,8 +7,8 @@ test_that("check_metadata_metabolites returns the right number of issues", {
 })
 
 test_that("check_metadata_sample returns the right number of issues", {
-  expect_equal(check_metadata_samples(df = metadata_sample_named, cas = "umichigan", return_n_issues = TRUE, verbose = FALSE), 3)
-  expect_equal(check_metadata_samples(df = metadata_sample_unnamed, cas = "umichigan", return_n_issues = TRUE, verbose = FALSE), 3)
+  expect_equal(check_metadata_samples(df = metadata_sample_named, cas = "umichigan", return_n_issues = TRUE, verbose = FALSE), 2)
+  expect_equal(check_metadata_samples(df = metadata_sample_unnamed, cas = "umichigan", return_n_issues = TRUE, verbose = FALSE), 2)
 })
 
 test_that("check_results returns the right number of issues", {