In bihealth/metaquac: Metabolomics Targeted Quality Control

The following QC analyses are based on the status-preprocessed dataset 1, unless explicitly stated otherwise.

# Select dataset for QC
biocrates <- datasets$discarded

# Get number of Sample, Pooled and Reference QCs
num_samples <- biocrates %>%
  select(Sample.Name, Sample.Type) %>% unique() %>%
  filter(Sample.Type == SAMPLE_TYPE_BIOLOGICAL) %>% nrow()
num_poolqc <- biocrates %>%
  select(Sample.Name, Sample.Type) %>% unique() %>%
  filter(Sample.Type == SAMPLE_TYPE_POOLED_QC) %>% nrow()
num_refqc <- biocrates %>%
  select(Sample.Name, Sample.Type) %>% unique() %>%
  filter(Sample.Type == ENV$SAMPLE_TYPE_REFERENCE_QC) %>% nrow()
num_cal <- biocrates %>%
  select(Sample.Name, Sample.Type) %>% unique() %>%
  filter(startsWith(Sample.Type, "Standard L")) %>% nrow()

# Get number of Compounds
num_compounds <- biocrates %>%
  select(Compound) %>% unique() %>% nrow()

# Normalization disabled for now
# Need to figure out how it fits in with current preprocessing structure and data
biocrates_unnormalized <- biocrates
if ("additional_normalization" %in% names(params) &&
    params$additional_normalization != "None") {
  cat("## Additional normalization {.tabset}\n")
  cat("### Info\n")
  cat(paste0(
    "Additional normalization applied on biological samples.\n\n",
    "* Normalization: **", params$additional_normalization, "**\n"))
  if (params$additional_normalization == "PQN"){
    reference_type <- ifelse("pqn_reference_type" %in% names(params),
                             params$pqn_reference_type, SAMPLE_TYPE_BIOLOGICAL
    )
    cat(paste0("* PQN Reference Sample Type: ", reference_type, "\n"))
    biocrates_normalized <- normalize_pqn(data = biocrates,
                                          target_type = SAMPLE_TYPE_BIOLOGICAL,
                                          reference_type = reference_type,
                                          target_values = ENV$CONCENTRATION)
    biocrates <- biocrates_normalized
    NORMALIZED <- TRUE
  }
  cat("\n")
  cat("Look at QQ-plots for a visualization of sample similarity in terms of
    concentration distribution. *Sorted* concentrations of biological samples
    are plotted against *sorted* concentrations of a reference sample
    (hence quantile vs quantile). Here, the reference sample is a median
    sample, based on the medians for each compound. Similarity in distributions
    would result in similar relations to the reference (i.e. linear) and
    should apply for most of the samples. Artifacts may occur due to
    missing values, which introduce shifts due to the sorting.\n\n")

  cat("### QQ-Plot Before\n")
  print(sample_qqplot(biocrates_unnormalized, scale_log10 = FALSE))
  cat("### QQ-Plot Before (log10)\n")
  print(sample_qqplot(biocrates_unnormalized, scale_log10 = TRUE))
  cat("### QQ-Plot After\n")
  print(sample_qqplot(biocrates_normalized, scale_log10 = FALSE))
  cat("### QQ-Plot After (log10)\n")
  print(sample_qqplot(biocrates_normalized, scale_log10 = TRUE))
}

Missing values

Per sample {.tabset}

Missing values are counted per sample, i.e. how many compounds were not measured in the sample at all or reliably enough (depending on status preprocessing). First, missing value counts are visualized depending on the total concentration measured in a sample (with Well Positions indicated for a few data points in low density areas).

# Scatter plot with number of missing values vs concentration (sum in sample),
# partly labeled with well position
plot_sample_na_intens_scatter(data = biocrates, shape = BATCH,
                              total_of = ENV$CONCENTRATION,
                              label = ENV$PLOT_SAMPLE_LABEL)

Furthermore, missing value counts are summarized by histograms (primarily) including visual separation and comparison of relevant sample types, batches (if applicable) or study variable groups (if indicated).

Per sample type

# Histogram of missing values per sample, separately for different sample types
data_na <- plot_na_histogram(data = biocrates,
                             group = "Sample.Name",
                             target = ENV$CONCENTRATION,
                             compare_key = "Sample.Type",
                             compare_values = c(SAMPLE_TYPE_BIOLOGICAL,
                                                ENV$SAMPLE_TYPE_REFERENCE_QC,
                                                SAMPLE_TYPE_POOLED_QC))
print(data_na$plot)

# Scatter plot with number of missing values vs concentration (sum in sample),
# partly labeled with well position
plot_sample_na_intens_scatter(data = biocrates, shape = BATCH,
                              total_of = ENV$CONCENTRATION,
                              sample_types = c(SAMPLE_TYPE_BIOLOGICAL,
                                               ENV$SAMPLE_TYPE_REFERENCE_QC,
                                               SAMPLE_TYPE_POOLED_QC),
                              label = ENV$PLOT_SAMPLE_LABEL)

# Missing values summary
data_na$data %>%
  group_by(Sample.Type) %>%
  summarize(`Mean(# MVs)` = mean(`# Missing Values`),
            `SD(# MVs)` = sd(`# Missing Values`),
            # `# Compounds` should be the same for all types, if all samples contain the same compounds
            `# Compounds` = mean(`# Total`),
            `# Samples` = n()) %>%
  easy_datatable(caption = "Missing values summary per sample type.", show_type = "statistics",
                 round_colums = c("Mean(# MVs)", "SD(# MVs)", "# Compounds"))

# Histograms of missing values per sample, separately for different batches
cat("\n#### In batches {.tabset}\n")
cat("\n\n##### All samples\n")
# MV histogram
data_na <- plot_na_histogram(data = biocrates,
                             group = "Sample.Name",
                             target = ENV$CONCENTRATION,
                             compare_key = BATCH,
                             compare_values = unique(biocrates[[BATCH]]))
print(data_na$plot)

# Scatter plot mv vs concentration
plot_sample_na_intens_scatter(data = biocrates, color = BATCH,
                              total_of = ENV$CONCENTRATION,
                              sample_types = NULL,
                              label = ENV$PLOT_SAMPLE_LABEL)

# Missing values summary
data_na$data %>%
  group_by(UQ(sym(BATCH))) %>%
  summarize(`Mean(# MVs)` = mean(`# Missing Values`),
            `SD(# MVs)` = sd(`# Missing Values`),
            # `# Compounds` should be the same for all batches, if all samples contain the same compounds
            `# Compounds` = mean(`# Total`),
            `# Samples` = n()) %>%
  easy_datatable(caption = "Missing values summary per batch.", show_type = "statistics",
                 round_colums = c("Mean(# MVs)", "SD(# MVs)", "# Compounds"))

cat("\n\n##### Reference QCs\n")
# MV histogram
data_na <- plot_na_histogram(data = biocrates %>% filter(Sample.Type == ENV$SAMPLE_TYPE_REFERENCE_QC),
                             group = "Sample.Name",
                             target = ENV$CONCENTRATION,
                             compare_key = BATCH,
                             compare_values = unique(biocrates[[BATCH]]))
print(data_na$plot)

# Scatter plot mv vs concentration
plot_sample_na_intens_scatter(data = biocrates, color = BATCH,
                              total_of = ENV$CONCENTRATION,
                              sample_types = c(ENV$SAMPLE_TYPE_REFERENCE_QC),
                              label = ENV$PLOT_SAMPLE_LABEL)

# Missing values summary
data_na$data %>%
  group_by(UQ(sym(BATCH))) %>%
  summarize(`Mean(# MVs)` = mean(`# Missing Values`),
            `SD(# MVs)` = sd(`# Missing Values`),
            # `# Compounds` should be the same for all batches, if all samples contain the same compounds
            `# Compounds` = mean(`# Total`),
            `# Samples` = n()) %>%
  easy_datatable(caption = "Missing values summary per batch.", show_type = "statistics",
                 round_colums = c("Mean(# MVs)", "SD(# MVs)", "# Compounds"))

cat(paste0("\n\n##### ", SAMPLE_TYPE_POOLED_QC, "s\n"))
# MV histogram
data_na <- plot_na_histogram(data = biocrates %>% filter(Sample.Type == SAMPLE_TYPE_POOLED_QC),
                             group = "Sample.Name",
                             target = ENV$CONCENTRATION,
                             compare_key = BATCH,
                             compare_values = unique(biocrates[[BATCH]]))
print(data_na$plot)

# Scatter plot mv vs concentration
plot_sample_na_intens_scatter(data = biocrates, color = BATCH,
                              total_of = ENV$CONCENTRATION,
                              sample_types = c(SAMPLE_TYPE_POOLED_QC),
                              label = ENV$PLOT_SAMPLE_LABEL)

# Missing values summary
data_na$data %>%
  group_by(UQ(sym(BATCH))) %>%
  summarize(`Mean(# MVs)` = mean(`# Missing Values`),
            `SD(# MVs)` = sd(`# Missing Values`),
            # `# Compounds` should be the same for all batches, if all samples contain the same compounds
            `# Compounds` = mean(`# Total`),
            `# Samples` = n()) %>%
  easy_datatable(caption = "Missing values summary per batch.", show_type = "statistics",
                 round_colums = c("Mean(# MVs)", "SD(# MVs)", "# Compounds"))

cat("\n\n##### Biological sample\n")
# MV histogram
data_na <- plot_na_histogram(data = biocrates %>% filter(Sample.Type == SAMPLE_TYPE_BIOLOGICAL),
                             group = "Sample.Name",
                             target = ENV$CONCENTRATION,
                             compare_key = BATCH,
                             compare_values = unique(biocrates[[BATCH]]))
print(data_na$plot)

# Scatter plot mv vs concentration
plot_sample_na_intens_scatter(data = biocrates, color = BATCH,
                              total_of = ENV$CONCENTRATION,
                              sample_types = c(SAMPLE_TYPE_BIOLOGICAL),
                              label = ENV$PLOT_SAMPLE_LABEL)

# Missing values summary
data_na$data %>%
  group_by(UQ(sym(BATCH))) %>%
  summarize(`Mean(# MVs)` = mean(`# Missing Values`),
            `SD(# MVs)` = sd(`# Missing Values`),
            # `# Compounds` should be the same for all batches, if all samples contain the same compounds
            `# Compounds` = mean(`# Total`),
            `# Samples` = n()) %>%
  easy_datatable(caption = "Missing values summary per batch.", show_type = "statistics",
                 round_colums = c("Mean(# MVs)", "SD(# MVs)", "# Compounds"))

# Compare sample missing values over different study variable groups
mv_plots_fun <- function(var, data, info, parent){

  # Histogram of missing values per sample,
  # separately for different study variable groups
  data_na <- plot_na_histogram(data = data,
                               group = "Sample.Name",
                               target = ENV$CONCENTRATION,
                               compare_key = var,
                               compare_values = unique(data[[var]]))

  # Scatter plot with number of missing values vs total intensity (sum in
  # sample), partly labeled with well position.
  plot2 <- plot_sample_na_intens_scatter(data = data,
                                         total_of = ENV$CONCENTRATION,
                                         color = var, label = ENV$PLOT_SAMPLE_LABEL,
                                         shape = BATCH)

  return(list(
    Header = paste0("\n##### ", info, var, "\n"),
    Plot1 = data_na$plot,
    Plot2 = plot2))
}

if (length(STUDY_VARIABLES) > 0){
  cat("#### Per study variable {.tabset}\n")

  var_plots <- recursive_execution(vars = STUDY_VARIABLES,
                                   end_fun = mv_plots_fun,
                                   data = biocrates %>% filter(Sample.Type == SAMPLE_TYPE_BIOLOGICAL))
  for (var_plot in var_plots){
    cat(var_plot$Header)
    print(var_plot$Plot1)
    print(var_plot$Plot2)
  }
}

Per compound {.tabset}

Missing values are counted per compound, i.e. how many samples don't feature a reliable measurement of the compound (depending on status preprocessing). Counts are calculated and visualized for different compound classes as well as for different sample subsets. I.e., for the same compound, missing values are separately counted within different sample types or within different study variable groups (if indicated).

Per compound class

# Histogram of missing values per compound, separately for different sample types
data_na <- plot_na_histogram(data = biocrates,
                             group = "Compound",
                             target = ENV$CONCENTRATION,
                             compare_key = "Class",
                             compare_values = unique(biocrates[["Class"]]))
print(data_na$plot)

Per sample type

# Histogram of missing values per compound, separately for different sample types
data_na <- plot_na_histogram(data = biocrates,
                             group = "Compound",
                             target = ENV$CONCENTRATION,
                             compare_key = "Sample.Type",
                             compare_values = c(SAMPLE_TYPE_BIOLOGICAL,
                                                ENV$SAMPLE_TYPE_REFERENCE_QC,
                                                SAMPLE_TYPE_POOLED_QC))
print(data_na$plot)

fig_width_mv_variable_bars <- 7
if (ENOUGH_POOLED_QC&&ENOUGH_REFERENCE_QC){
  compounds_with_mvs <- biocrates %>%
    filter(
      !!sym(COLUMN_SAMPLE_TYPE) %in%
        c(SAMPLE_TYPE_POOLED_QC, ENV$SAMPLE_TYPE_REFERENCE_QC)
    ) %>%
    group_by(Compound) %>%
    filter(na_percent(UQ(sym(ENV$CONCENTRATION))) > 0) %>%
    select(Compound) %>%
    distinct()
  fig_width_mv_variable_bars <- max(5, nrow(compounds_with_mvs) * 0.11 + 0.45)
}

cat("\n#### Per QC type \n")

cat("
The following figure compares the absolute and relative numbers of missing
values (MVs) per compound for reference and pooled QC samples. This comparison
can serve as a basic global indicator whether the origin of a missing value is
of technical or biological nature.
Low numbers of MVs in reference QC samples indicate technically measurable
compounds. In this case, a high number of MVs in pooled QC samples suggests that
a compound is biologically missing. However, it must be noted that MVs in pool
and biological samples could also be a result of sample treatment and
extraction. In the case of high numbers of MVs in reference QC samples, no
assumption can be made about MVs in pool and biological samples. They might only
result from technical difficulties in the assay (such as with the MVs in the
reference QC samples), but could also have been missing before (either
biologically or due to treatment and extraction).
Only compounds actually featuring missing values are shown in this figure.
    ")

qc_comp_plot <- plot_compound_na_variable_bars(
  data = biocrates,
  compare = COLUMN_SAMPLE_TYPE,
  target = ENV$CONCENTRATION,
  sample_type = c(SAMPLE_TYPE_POOLED_QC, ENV$SAMPLE_TYPE_REFERENCE_QC),
  add_norm_perc_mvs = FALSE,
  position = "dodge"
)

qc_comp_data <- calc_compound_na_variable(
  data = biocrates,
  compare = COLUMN_SAMPLE_TYPE,
  target = ENV$CONCENTRATION,
  sample_type = c(SAMPLE_TYPE_POOLED_QC, ENV$SAMPLE_TYPE_REFERENCE_QC),
  add_norm_perc_mvs = FALSE
)

print(qc_comp_plot)
cat("\n")
easy_datatable(
  qc_comp_data,
  caption = "Comparison of Pooled and Reference MVs.",
  show_type = "statistics"
)
cat("\n")

compounds_with_mvs <- biocrates %>%
  filter(Sample.Type == SAMPLE_TYPE_BIOLOGICAL) %>%
  group_by(Compound) %>%
  filter(na_percent(UQ(sym(ENV$CONCENTRATION))) < 100,
         na_percent(UQ(sym(ENV$CONCENTRATION))) > 0) %>%
  select(Compound) %>%
  distinct()
fig_width_mv_variable_bars <- max(5, nrow(compounds_with_mvs) * 0.11 + 0.45)

# Compare compound missing values over different study variable groups
mv_plots_fun <- function(var, data, info, parent){

  plot1 <- plot_compound_na_variable_bars(
    data = data,
    compare = var,
    target = ENV$CONCENTRATION,
    sample_type = SAMPLE_TYPE_BIOLOGICAL
  )

  data1 <- calc_compound_na_variable(
    data = data,
    compare = var,
    target = ENV$CONCENTRATION,
    sample_type = SAMPLE_TYPE_BIOLOGICAL
  )

  return(list(
    Header = paste0("\n##### ", info, var, "\n"),
    Plot1 = plot1,
    Data1 = data1))
}

if (length(STUDY_VARIABLES) > 0){
  cat("\n#### Per study variable {.tabset}\n")
  cat("\nBar plots per study variable visualize differences in the occurrence of missing
      values in different study groups. **# Missing Values** counts are used as reference to
      show whether a compound is missing just a bit or a lot. It shouldn't be used for
      comparison of groups due to possible differences in group sizes. However, **% Missing
      Values** show the percentage of missing values per group, which are further added
      up and normalized to one as **Normalized % Missing Values**. This allows to
      infer compounds which are considerably more missing in certain groups (if the
      missing value count is high). Only compounds with a least one and less than 100%
      missing values (overall) are shown.\n")
  var_plots <- recursive_execution(vars = STUDY_VARIABLES,
                                   end_fun = mv_plots_fun,
                                   data = biocrates %>% filter(Sample.Type == SAMPLE_TYPE_BIOLOGICAL))
  for (var_plot in var_plots) {
    cat(var_plot$Header)
    print(var_plot$Plot1)
    cat("\n")
    cat(knitr::knit_print(
      easy_datatable(var_plot$Data1, caption = "Compound MVs within study groups.",
                     show_type = "statistics")))
    cat("\n")
  }
}

Sample variability

Sequence progression {.tabset}

Progression of sample summary statistics (totals for concentration/area/intensity and missing values) is visualized over the acquisition sequence. This allows an quick overview of variability between replicate samples as well as between batches, while the parallel view on the different data types (concentration, area, istd area, etc.) illustrates the effect of calibration/normalization.

After calibration (concentration), replicate samples (such as reference and pooled QCs) should feature a preferably horizontal regression line (in particular for primarily multi-point calibrated data). Furthermore, regression lines of different batches should align reasonably to justify subsequent joint analysis. In comparison, uncalibrated and unnormalized measurements (area and intensity) may feature incomparable fluctuation and batch drifts.

message("**Note**: Many LC compounds in the Biocrates MxP Quant 500 Kit are 1-point calibrated and
don't feature an external standard. Thus, a high number of missing values is to be expected within
standard calibration samples after preprocessing according to status (mainly due to \"< LOD\").
For the same reason, standard calibration samples are not expected to encompass all the other
samples with respect to totals of concentration, area or intensity. For instance, the Biocrates
QC Level 3 sample should feature higher totals than the highest standard calibration samples.")

fig_height_sequence <- length(DATA_TYPES) * 1.5 + 2

Totals overview

plot_sequence_overview_total(biocrates, targets = DATA_TYPES)

Totals details {.tabset}

Linear models are calculated to assess the horizontal behavior of measurements (total concentration per sample) within specific sample types. In particular, in a perfect batch with no deviation the resulting regression lines should be horizontal. Statistical analysis is applied to highlight potential deviation of the regression line fitted to the experimental data from the horizontal.

For single batches, the simple linear model Total Concentration ~ Sequence Position is checked for unwanted batch drifts inferred from significant slopes. Significant slopes may not only indicate batch drifts but general technical variability not sufficiently resolved by internal standard normalization and external standard calibration.

For batch comparison, the model Total Concentration ~ Batch + Sequence Position / Batch is used to infer unwanted significant differences in intercepts and slopes between batches (with one batch randomly selected as reference). Furthermore, a Kruskal-Wallis rank sum test is used to infer unwanted general differences in batch distributions.

message("**Note**: Use the automated rating for guidance only and in addition judge manually (e.g.
by visual inspection of the totals overview) whether total concentration fits behave well enough in
comparison to area or intensity (i.e. unnormalized and uncalibrated data). The rating remains rather
subjective for now, as it will require a great number of experiments and extensive validation to
assess which model coefficients and p-values are acceptable to support reliable analysis.")

message("**Note**: Expect inferior results for FIA data due to one-point calibration (in contrast to
7-point calibration in LC data).")

cat(paste0("\n\n##### Reference QCs {.tabset}\n"))
cat("\n\n###### Single batches\n")
easy_datatable(sequence_batch_horizontality(data = biocrates, sample_type = ENV$SAMPLE_TYPE_REFERENCE_QC),
               caption = "Single batch models coefficients.", show_type = "statistics")

cat("\n\n###### Batch comparison\n")
results <- sequence_batch_conformity(
  data = biocrates,
  sample_type = ENV$SAMPLE_TYPE_REFERENCE_QC
)
cat(paste("\n\nKruskal-Wallis p-value:", results$kruskal_pvalue, "\n"))
cat(paste("\n\nModel fit:", results$r_squared, "\n"))
easy_datatable(results$coefficients, caption = "Multi batch model coefficients.",
               show_type = "statistics")

cat("\n\n###### Auxiliary Figures\n")
assisting_plots <- sequence_batch_assistant_plots(data = biocrates, sample_type = ENV$SAMPLE_TYPE_REFERENCE_QC)
print(assisting_plots$sequence)
print(assisting_plots$distribution)

cat(paste0("\n\n##### ", SAMPLE_TYPE_POOLED_QC, "s {.tabset}\n"))
cat("\n\n###### Single batches\n")
easy_datatable(sequence_batch_horizontality(data = biocrates, sample_type = SAMPLE_TYPE_POOLED_QC),
               caption = "Single batch models coefficients.", show_type = "statistics")

cat("\n\n###### Batch comparison\n")
results <- sequence_batch_conformity(
  data = biocrates,
  sample_type = SAMPLE_TYPE_POOLED_QC
)
cat(paste("\n\nKruskal-Wallis p-value:", results$kruskal_pvalue, "\n"))
cat(paste("\n\nModel fit:", results$r_squared, "\n"))
easy_datatable(results$coefficients, caption = "Multi batch model coefficients.",
               show_type = "statistics")

cat("\n\n###### Auxiliary Figures\n")
assisting_plots <- sequence_batch_assistant_plots(data = biocrates, sample_type = SAMPLE_TYPE_POOLED_QC)
print(assisting_plots$sequence)
print(assisting_plots$distribution)

Biological samples {.tabset}

cat("\n\n###### Single batches\n")
easy_datatable(sequence_batch_horizontality(data = biocrates, sample_type = SAMPLE_TYPE_BIOLOGICAL),
               caption = "Single batch models coefficients.", show_type = "statistics")

cat("\n\n###### Batch comparison\n")
results <- sequence_batch_conformity(data = biocrates, sample_type = SAMPLE_TYPE_BIOLOGICAL)
cat(paste("\n\nKruskal-Wallis p-value:", results$kruskal_pvalue, "\n"))
cat(paste("\n\nModel fit:", results$r_squared, "\n"))
easy_datatable(results$coefficients, caption = "Multi batch model coefficients.",
               show_type = "statistics")

cat("\n\n###### Auxiliary Figures\n")
assisting_plots <- sequence_batch_assistant_plots(data = biocrates, sample_type = SAMPLE_TYPE_BIOLOGICAL)
print(assisting_plots$sequence)
print(assisting_plots$distribution)

MVs

plot_sequence_overview_mv(biocrates, targets = DATA_TYPES)

Well Plate Overview {.tabset}

Sample summary statistics (missing values and total concentration/area/intensity) are compared with respect to well plate position. This allows to identify unexpected position-based issues (e.g. edge-effects), as might be demonstrated by a cluster of aberrating dots, or to spot outlier samples. In general, samples of the same type should feature similar dot sizes (with minor variations in biological samples). Samples areas and intensities will vary more than concentrations (as latter are calibrated).

fig_height_well_plate <- ceiling(length(unique(biocrates$Batch)) / 2) * 3 + 2

# Are well plate positions available?
HAS_WELL_POSITION <- COLUMN_WELL_POSITION %in% names(biocrates)

message("**Note**: Well Position not available.")

if ("CONCENTRATION" %in% names(DATA_TYPES)){
  cat("#### % Missing Values of Concentration\n")
  plot_well_plate_overview_mv(biocrates, target = ENV$CONCENTRATION)
}

if ("CONCENTRATION" %in% names(DATA_TYPES)){
  cat("#### Total Concentration\n")
  plot_well_plate_overview_total(biocrates, target = ENV$CONCENTRATION)
}

if ("AREA" %in% names(DATA_TYPES)){
  cat("#### Total Area")
  plot_well_plate_overview_total(biocrates, target = ENV$AREA)
}

if ("INTENSITY" %in% names(DATA_TYPES)){
  cat("#### Total Intensity\n")
  plot_well_plate_overview_total(biocrates, target = ENV$INTENSITY)
}

if ("ISTD_AREA" %in% names(DATA_TYPES)){
  cat("\n#### Total Area ISTD\n")
  plot_well_plate_overview_total(biocrates, target = ENV$ISTD_AREA)
}

if ("ISTD_INTENSITY" %in% names(DATA_TYPES)){
  cat("\n#### Total Intensity ISTD\n")
  plot_well_plate_overview_total(biocrates, target = ENV$ISTD_INTENSITY)
}

Samples overlapping by type {.tabset}

Violin and box plots summarize compound measurements per sample and are grouped and overlapped per sample type, if applicable. This allows an quick overview of variability between replicate samples such as reference and pooled QCs as well as biological samples, while the parallel view on the different data types illustrates the effect of calibration/normalization (i.e. replicate sample concentration profiles should line up rather well).

num_sample_types <- length(unique(biocrates$Sample.Type))
num_batches <- length(unique(biocrates$Batch))
fig_height_variability_overlap <- min(num_sample_types * num_batches * 0.25 + 1.5, 10)

Violins

# Measurement violins per sample, overlapping per sample type
plot_sample_variability_overlapped(biocrates,
                                   targets = DATA_TYPES,
                                   sample_types = c(SAMPLE_TYPE_BIOLOGICAL,
                                                    ENV$SAMPLE_TYPE_REFERENCE_QC,
                                                    SAMPLE_TYPE_POOLED_QC,
                                                    paste0("Standard L", 1:7)),
                                   plot_type = "violin",
                                   log10 = TRUE)

Boxplots

# Measurement boxplots per sample, overlapping per sample type
plot_sample_variability_overlapped(biocrates,
                                   targets = DATA_TYPES,
                                   sample_types = c(SAMPLE_TYPE_BIOLOGICAL,
                                                    ENV$SAMPLE_TYPE_REFERENCE_QC,
                                                    SAMPLE_TYPE_POOLED_QC,
                                                    paste0("Standard L", 1:7)),
                                   plot_type = "boxplot",
                                   log10 = TRUE)

QC samples ordered by aquisition {.tabset}

Violin and box plots summarize compound measurements per replicate sample, grouped per sample type and ordered according to acquisition sequence. This allows a quick overview of variability progression of replicate samples such as Reference or Pooled QC samples over the batch (e.g. to indicate batch drift), while the parallel view on the different data types illustrates the effect of calibration/normalization (i.e. replicate sample concentration profiles should line up rather well).

fig_height_variability_ordered <- length(unique(biocrates$Batch)) * 2 + 1

if (!(AVAILABLE_REFERENCE_QC || AVAILABLE_REFERENCE_QC)){
  warn_text <- paste0(
    "Warning: No QC samples available (Reference or Pooled QC)."
  )
  message(warn_text)
}

# Measurement violins per sample, overlapping per sample type
cat("\n#### Violins\n")
plot_sample_variability_sequential(biocrates,
                                   targets = DATA_TYPES,
                                   sample_types = c(ENV$SAMPLE_TYPE_REFERENCE_QC,
                                                    SAMPLE_TYPE_POOLED_QC),
                                   plot_type = "violin",
                                   log10 = TRUE)

# Measurement boxplots per sample, overlapping per sample type
cat("\n#### Boxplot\n")
plot_sample_variability_sequential(biocrates,
                                   targets = DATA_TYPES,
                                   sample_types = c(ENV$SAMPLE_TYPE_REFERENCE_QC,
                                                    SAMPLE_TYPE_POOLED_QC),
                                   plot_type = "boxplot",
                                   log10 = TRUE)

# Remove "biocrates" dataset to ensure following sections select the dataset they need
rm(biocrates)

bihealth/metaquac documentation built on Aug. 7, 2021, 8:40 a.m.