Merge pull request #204 from MoTrPAC/develop

v0.8.7: Metabolomics batching variables QC support

Author: biodavidjm

.circleci/config.yml

@@ -1,10 +1,36 @@
-version: 1
+version: 2.1
 jobs:
-  build:
+  build_4_2_2:
     docker:
-      - image: rocker/verse:3.6.2
+      - image: rocker/verse:4.2.2
     steps:
       - checkout
+      - run:
+          name: Install LaTeX packages
+          command: |
+            apt-get update
+            apt-get install -y texlive
+            apt-get install -y texlive-fonts-recommended
+      - run:
+          name: Install package dependencies
+          command: R -e "devtools::install_deps(dep = TRUE)"
+      - run:
+          name: Build package
+          command: R CMD build .
+      - run:
+          name: Check package
+          command: R CMD check *tar.gz
+  
+  build_4_1_0:
+    docker:
+      - image: rocker/verse:4.1.0
+    steps:
+      - checkout
+      - run:
+          name: Install LaTeX packages
+          command: |
+            apt-get update
+            apt-get install -y texlive-fonts-recommended
       - run:
           name: Install package dependencies
           command: R -e "devtools::install_deps(dep = TRUE)"
@@ -13,4 +39,11 @@ jobs:
           command: R CMD build .
       - run:
           name: Check package
-          command: R CMD check *tar.gz
+          command: R CMD check *tar.gz
+
+workflows:
+  version: 2
+  build_and_test:
+    jobs:
+      - build_4_2_2
+      - build_4_1_0

DESCRIPTION

@@ -1,8 +1,8 @@
 Package: MotrpacBicQC
 Type: Package
 Title: QC/QA functions for the MoTrPAC community
-Version: 0.8.6
-Date: 2023-04-21
+Version: 0.8.7
+Date: 2023-05-22
 Author: MoTrPAC Bioinformatics Center
 Maintainer: David Jimenez-Morales <[email protected]>
 Description: R Package for the analysis of MoTrPAC datasets. 
@@ -19,7 +19,7 @@ URL: https://github.com/MoTrPAC/MotrpacBicQC
 Encoding: UTF-8
 LazyData: true
 BugReports: https://github.com/MoTrPAC/MotrpacBicQC/issues
-Depends: R (>= 3.6.0)
+Depends: R (>= 4.1.0)
 Imports:
     data.table,
     dplyr,
@@ -30,6 +30,7 @@ Imports:
     inspectdf,
     jsonlite,
     knitr,
+    lubridate,
     naniar,
     progress,
     purrr,

NAMESPACE

@@ -31,13 +31,17 @@ export(set_phase)
 export(validate_assay)
 export(validate_batch)
 export(validate_cas)
+export(validate_dates_times)
+export(validate_lc_column_id)
 export(validate_metabolomics)
+export(validate_na_empty)
 export(validate_phase)
 export(validate_processFolder)
 export(validate_proteomics)
 export(validate_refmetname)
 export(validate_tissue)
 export(validate_two_phases)
+export(validate_yyyymmdd_dates)
 export(write_metabolomics_releases)
 export(write_proteomics_releases)
 import(dplyr)
@@ -60,6 +64,7 @@ importFrom(gridExtra,arrangeGrob)
 importFrom(gridExtra,grid.arrange)
 importFrom(inspectdf,inspect_na)
 importFrom(jsonlite,fromJSON)
+importFrom(lubridate,parse_date_time)
 importFrom(readr,read_lines)
 importFrom(scales,percent)
 importFrom(stats,median)

NEWS.md

@@ -1,4 +1,11 @@
-# MotrpacBicQC 0.8.5 (2023-04-21)
+# MotrpacBicQC 0.8.7 (2023-05-22)
+
+* Add QC for the new required batching variables
+* Replace deprecated `ggplot` function
+* Fix issues with `dl_read_gcp`
+* Other adjustments
+
+# MotrpacBicQC 0.8.6 (2023-04-21)
 
 * Minor adjustments
 

R/metabolomics_plots.R

@@ -54,7 +54,7 @@ plot_basic_metabolomics_qc <- function(results,
   if(verbose) message("       - (p) Density distributions")
   mu <- results_long %>%
     group_by(sample_id) %>%
-    reframe(grp.mean = mean(intensity))
+    dplyr::reframe(grp.mean = mean(intensity))
 
   den1 <- ggplot(data = results_long, aes(x = log2(intensity), color = sample_type)) + 
     geom_density(na.rm = TRUE) + 
@@ -83,7 +83,7 @@ plot_basic_metabolomics_qc <- function(results,
   if(verbose) message("       - (p) Plot sum of intensity/concentration values")
   sum_int <- results_long %>%
     group_by(sample_id, sample_type, sample_order) %>%
-    reframe(sum_quant = sum(intensity))
+    dplyr::reframe(sum_quant = sum(intensity))
 
   psumint <- ggplot(sum_int, aes(x = reorder(sample_id, sample_order), y = sum_quant, fill = sample_type)) +
     geom_bar(stat = "identity") + theme_classic() +
@@ -178,7 +178,7 @@ plot_basic_metabolomics_qc <- function(results,
   if(verbose) message("       - (p) Plot ID counts")
   uid <- results_long %>%
     group_by(across(all_of(c("metabolite_name", "sample_id", "sample_type", "sample_order", "id_type")))) %>%
-    reframe(total_intensity = intensity)
+    dplyr::reframe(total_intensity = intensity)
   uid2 <- uid[which(!is.na(uid$total_intensity)),]
   uid3 <- unique(uid2[c("metabolite_name", "sample_id", "sample_type", "sample_order", "id_type")]) %>%
     count(sample_id, sample_type, sample_order, id_type)
@@ -268,9 +268,9 @@ plot_basic_metabolomics_qc <- function(results,
 
 
   ppids <- ggplot(results, aes(x = as.factor(id_type), fill = as.factor(id_type))) +
-    geom_bar(aes(y = (..count..)/sum(..count..))) +
-    geom_text(aes(y = ((..count..)/sum(..count..)), 
-                  label = scales::percent((..count..)/sum(..count..))), 
+    geom_bar(aes(y = after_stat(count)/sum(after_stat(count)))) +
+    geom_text(aes(y = after_stat(count)/sum(after_stat(count)), 
+                  label = scales::percent(after_stat(count)/sum(after_stat(count)))), 
               stat = "count", vjust = -0.25) +
     scale_y_continuous(labels = percent) +
     labs(title = "Proportion of Features Identified (named/unnamed)", 
@@ -283,7 +283,7 @@ plot_basic_metabolomics_qc <- function(results,
 
   pnids <- ggplot(results, aes(x = id_type, fill = id_type)) +
     geom_bar() +
-    geom_text(stat = 'count',aes(label =..count.., vjust = -0.2)) +
+    geom_text(stat = 'count',aes(label = after_stat(count), vjust = -0.2)) +
     labs(title = "Total Number of Features Identified (named/unnamed)", 
          subtitle = paste(output_prefix), y = "", x = "") +
     theme_light() +

R/metabolomics_qc.R

@@ -3,8 +3,7 @@
 utils::globalVariables(
   c("assay_codes",
     "bic_animal_tissue_code",
-    "phenotypes_pass1a06_short",
-    "..count.."))
+    "phenotypes_pass1a06_short"))
 
 
 # METABOLOMICS DATASETS: PRIMARY QC
@@ -72,7 +71,7 @@ check_metadata_metabolites <- function(df,
         if(verbose) message("  + (+) `refmet_name` unique values: OK")
       }
 
-      if(verbose) message("  + Validating `refmet_name`")
+      if(verbose) message("  + Validating `refmet_name` (it might take some time)")
       nrnna <- validate_refmetname(dataf = df, verbose = verbose)
       if(nrnna > 0){
         if(verbose) message(paste0("   - (-) SUMMARY: ", nrnna, " `refmet_name` not found in RefMet Metabolomics Data Dictionary: FAIL"))
@@ -184,7 +183,7 @@ check_metadata_samples <- function(df,
   # filter only expected columns
   df <- filter_required_columns(df = df,
                                 type = "m_s",
-                                verbose = FALSE)
+                                verbose = TRUE)
 
   # Check every column
   # sample_id: si
@@ -264,6 +263,40 @@ check_metadata_samples <- function(df,
     if(verbose) message("   - (-) `raw_file` column missed: FAIL")
     ic <- ic + 1
   }
+  
+  if("extraction_date" %in% colnames(df)){
+    if(any(is.na(df$extraction_date))){
+      if(verbose) message("   - (-) `extraction_date` has NA values: FAIL")
+      ic <- ic + 1
+    }else{
+      icdate <- validate_yyyymmdd_dates(df = df, date_column = "extraction_date", verbose = verbose)
+      ic <- ic + icdate
+    }
+  }else{
+    if(verbose) message("   - (-) `extraction_date` column missed: FAIL")
+    ic <- ic + 1
+  }
+  
+  if("acquisition_date" %in% colnames(df)){
+    if( any(grepl(":", df$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)
+    }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")
+    ic <- ic + 1
+  }
+  
+  # check if lc_column_id is in column names
+  if ("lc_column_id" %in% colnames(df)) {
+    validate_lc_column_id(df, column_name = "lc_column_id", verbose = verbose)
+  }else{
+    if(verbose) message("   - (-) `lc_column_id` column missed: FAIL")
+    ic <- ic + 1
+  }
 
   if(return_n_issues) return(ic)
 } #end check_metadata_samples
@@ -658,7 +691,7 @@ validate_metabolomics <- function(input_results_folder,
       if(isTRUE(all.equal(m_s_n, m_s_u))){
         if(verbose) message("  + (+) Metadata samples: named and unnamed are identical: OK")
       }else{
-        if(verbose) message("   - (-) Metadata samples: named and unnamed files differ")
+        if(verbose) message("   - (-) Metadata samples: named and unnamed files differ: FAIL")
         ic <- ic + 1
       }
     }else{