Package 'CalibraCurve'

Title: Calibration curves for targeted proteomics, lipidomics and metabolomics data
Description: CalibraCurve is a computational tool designed to generate calibration curves for targeted mass spectrometry-based quantitative data. It is applicable to various omics disciplines, including proteomics, lipidomics, and metabolomics. The package also offers functionalities for data and calibration curve visualization and concentration prediction from new datasets based on the established curves.
Authors: Karin Schork [aut, cre] (ORCID: <https://orcid.org/0000-0003-3756-4347>), Robin Grugel [aut], Michael Kohl [aut], Markus Stepath [aut], Martin Eisenacher [aut, fnd]
Maintainer: Karin Schork <[email protected]>
License: BSD 3-clause License + file LICENSE
Version: 1.3.0
Built: 2026-05-31 06:41:28 UTC
Source: https://github.com/bioc/CalibraCurve

Help Index

PLR: Calculate key features for the existing continuous preliminary ranges

Description

PLR: Calculate key features for the existing continuous preliminary ranges

Usage

.calcContPrelimRanges(index)

Arguments

index

integer
Integer vector containing indices of concentration levels with CV < threshold.

Value

Data.frame with one range in each row. Contains the columns "startPoints", "endPoints", and "extent".

PLR: calculate CV for one concentration level

Description

PLR: calculate CV for one concentration level

Usage

.calcCV(x)

Arguments

x

data.frame
Data.frame containing data for a specific concentration level.

Value

numeric(1)
Coefficient of variation (CV) for the given concentration level.

FLR: calculate weighted or unweighted linear model

Description

FLR: calculate weighted or unweighted linear model

Usage

.calcLinearModel(x, weights = NULL)

Arguments

x

list of data.frames
List of data frames (one entry for each concentration level), e.g. output "dataPrelim" from calculate_PLR.
weights

numeric
Vector of weights as calculated by applying .calcWeights (default is NULL and will result in an unweighted model).

Value

Fit of the linear model as an object of class "lm".

FLR: calculate percent bias for a specific concentration level

Description

FLR: calculate percent bias for a specific concentration level

Usage

.calcPerBias(x, LMfit, expConc)

Arguments

x

data.frame
Data.frame containing data for a specific concentration level.
LMfit

lm object
Linear model fit as calculated by .calcLinearModel.
expConc

numeric(1)
Expected (known) concentration level.

Value

Vector of percent bias values for each data point in this concentration level.

FLR: calculate average, SD and CV percent bias for each concentration level

Description

FLR: calculate average, SD and CV percent bias for each concentration level

Usage

.calcPerBiasAvgSDCV(x, method = "mean")

Arguments

x

list
Result of .calcPerBiasLevels.
method

character(1)
Method for calculating the average percent bias: "mean" (default) or "median".

Value

data frame with 3 columns: avgPerBias, stdDevPerBias, CV_PerBias
each row is one concentration level

FLR: calculate list of percent bias values for all concentration levels

Description

FLR: calculate list of percent bias values for all concentration levels

Usage

.calcPerBiasLevels(x, LMfit)

Arguments

x

list of data.frames
List of data frames (one entry for each concentration level), e.g. output "dataPrelim" from calculate_PLR.
LMfit

lm object
Linear model fit as calculated by .calcLinearModel.

Value

List with percent bias values for each data point per concentration level

Calculate Response factors

Description

Function, which calculates a response factor for a single data point

Usage

.calcResponseFactors(x, intercept, expConc)

Arguments

x

data.frame*
Data.frame containing data for a specific concentration level.
intercept

numeric(1)
Intercept of the linear model.
expConc

numeric(1)
Expected concentration (known concentration value).

Details

Formula obtained from: Green, J. M., A practical guide to analytical method validation. Analytical Chemistry 1996, 68, 305A-309A.

Value

vector of response factors for this specific concentration level

FLR: calculate weights for linear model for a specific concentration level

Description

FLR: calculate weights for linear model for a specific concentration level

Usage

.calcWeights(x, weightingMethod = "1/x^2")

Arguments

x

data.frame
Data.frame containing data for a specific concentration level.
weightingMethod

character(1)
Method for weighting (currently "1/x", "1/x^2" and "None" are supported, default is "1/x^2").

Value

Vector of a constant weights for each measurement in this concentration level.

FLR: checks if final linear range has been reached (compare average percent bias with threshold)

Description

FLR: checks if final linear range has been reached (compare average percent bias with threshold)

Usage

.checkFinalRange(perBiasInfo, perBiasThres = 20)

Arguments

perBiasInfo

data.frame Result of .calcPerBiasAvgSDCV.
perBiasThres

numeric(1) numeric(1)
Threshold for average percent bias in percent, default is 20.

Value

TRUE if both lowest and highest concentration level passed the check (and the final linear range is reached), FALSE otherwise

Data preprocessing: Helper function to check for sufficient number of replicates for a specific concentration level

Description

Data preprocessing: Helper function to check for sufficient number of replicates for a specific concentration level

Usage

.checkNumberReplicates(x, data, minReplicates)

Arguments

x

numeric(1)
concentration level to check
data

list of data.frames
list of data.frames (each dataframe contains data for a specific concentration level)
minReplicates

integer(1)
minimal number of data points per concentration level

Value

logical(1)
TRUE if there are enough replicates, else FALSE

Data preprocessing: Helper function to select all rows from a specific concentration level

Description

Data preprocessing: Helper function to select all rows from a specific concentration level

Usage

.filterConcentrationLevel(x, data)

Arguments

x

numeric(1)
concentration level to select
data

data.frame
data set to be filtered (e.g., result of readDataTable)

Value

data.frame

Prepare annotation data

Description

Prepare annotation data

Usage

.prepareAnnotationData(D_calib)

Arguments

D_calib

data.frame
Result of .prepareCalibData

Value

Dataframe with annotation data for the plot.

Prepare calibration data

Description

Prepare calibration data

Usage

.prepareCalibData(CC_RES)

Arguments

CC_RES

list Result object of CalibraCurve.

Value

Dataframe with calibration data for plotting.

Prepare curve data

Description

Prepare curve data

Usage

.prepareCurveData(CC_RES, D_calib)

Arguments

CC_RES

list
Results of CalibraCurve. Each list element is the result of calc_single_curve
D_calib

data.frame
Result of .prepareCalibData

Value

Dataframe with data points on a grid to plot the calibration curve.

Save results of CalibraCurve

Description

Save results of CalibraCurve

Usage

.saveCCResult(CC_res, output_path, suffix = "")

Arguments

CC_res

list Result object of CalibraCurve.
output_path

character(1)
Path to the output directory.
suffix

character(1)
Suffix for the output files, ideally starting with "_" (default is "").

Value

Returns nothing, but the function saves the results to the specified output path.

Save result tables and plots

Description

Save result tables and plots

Usage

.saveTablesAndPlots(
  RES,
  output_path,
  pl_CC_list,
  pl_RF_list,
  summary_tab,
  CC_plot_width,
  CC_plot_height,
  RF_plot_width,
  RF_plot_height,
  plot_type,
  plot_dpi,
  device
)

Arguments

RES

list
Results of CalibraCurve. Each list element is the result of calc_single_curve
output_path

character(1)
Folder to save results (table and plots). If NULL (default), results are not saved.
pl_CC_list

list
List of calibration curves (ggplot objects, results of plotCalibraCurve (CC_plot)).
pl_RF_list

list
List of response factor plots (ggplot objects, results of plotResponseFactors).
summary_tab

data.frame
Table with summary information, result of plotCalibraCurve (annotation_dat))
CC_plot_width

numeric(1)
Plot width in cm (default is 10).
CC_plot_height

numeric(1)
Plot height in cm (default is 10).
RF_plot_width

numeric(1)
Plot width in cm (default is 10).
RF_plot_height

numeric(1)
Plot height in cm (default is 10).
plot_type

character(1)
Type of plot for calibration curves: "single_plots" (default, generate a separate plot for each substance), "multiplot" (generate a graphic with subplots for each substance) or "all_in_one" (generate a single plot with all substances).
plot_dpi

numeric(1)
Plot resolution in dpi (default is 300).
device

character(1)
Device for saving the plot (default is "png"). Other options include "pdf", "jpeg", "tiff", "svg" etc. For details see ggsave.

Value

Invisible NULL. Plots and the summary table are saved to the hard drive.

FLR: selects the highest or lowest concentration level for removal

Description

FLR: selects the highest or lowest concentration level for removal

Usage

.selctConcLevel(x, perBiasT = 20, consPerBiasCV = TRUE, perBiasDistT = 10)

Arguments

x

data.frame Result of .calcPerBiasAvgSDCV.
perBiasT

numeric(1)
Threshold for average percent bias in percent, default is 20.
consPerBiasCV

consider CV? default is TRUE
perBiasDistT

numeric(1)
Threshold for the difference in average percent bias in percent (for lower differences, CV will be considered), default is 10. Only used if consPerBiasCV is TRUE.

Value

TRUE, if lowest concentration will be removed, FALSE if highest will be removed

Assemble result tables

Description

Assemble result tables

Usage

assemble_results(
  X,
  dataCleaned,
  PLR_res,
  resFacDataV,
  avgResFacDataV,
  FLR_res,
  mod,
  cvThres = 20,
  RfThresL = 80,
  RfThresU = 120,
  substance = "substance1"
)

Arguments

X

data.frame
Original data set, e.g. result of readDataTable.
dataCleaned

list of data.frames
Cleaned data, result of cleanData.
PLR_res

list
Result object of calculate_PLR.
resFacDataV

list
List of response factor values for each concentration level.
avgResFacDataV

numeric
Vector of mean response factor values for the different concentration levels.
FLR_res

list Result object of calculate_FLR.
mod

lm object
Final linear model fit (object "mod" from results of calculate_FLR).
cvThres

numeric(1)
Threshold for CV per concentration level in percent (default is 20).
RfThresL

numeric(1)
Lower threshold for response factor in percent (default is 80).
RfThresU

numeric(1)
Upper threshold for response factor in percent (default is 120).
substance

character(1)
Name of the substance (default is "substance1").

Value

List with the following elements:

  • result_table_conc_levels: Result table with one line for each concentration level.

  • result_table_obs: Result table with one line per observation (e.g. individual response factors for each data point).

Examples

file <- system.file("extdata", "MSQC1", "msqc1_dil_GGPFSDSYR.rds",
package = "CalibraCurve")
D_list <- readDataSE(file, concColName = "amount_fmol",
        substColName = "Substance", assayNumber = 1)
data_cleaned <- cleanData(D_list[[1]])

RES_PLR <- calculate_PLR(data_cleaned, calcContinuousPrelimRanges = FALSE)
RES_FLR <- calculate_FLR(RES_PLR$dataPrelim)
RFs <- calcRF(data_cleaned, mod = RES_FLR$mod)

assemble_results(
    X = D_list[[1]],
    dataCleaned = data_cleaned,
    PLR_res = RES_PLR,
    resFacDataV = RFs$RFs,
    avgResFacDataV = RFs$meanRFs,
    FLR_res = RES_FLR,
    mod = RES_FLR$mod
)

Calculate all necessary steps for a single calibration curve

Description

Calculate all necessary steps for a single calibration curve

Usage

calc_single_curve(
  D,
  substance = "substance",
  minReplicates = 3,
  cvThres = 20,
  calcContinuousPrelimRanges = FALSE,
  weightingMethod = "1/x^2",
  centralTendencyMeasure = "mean",
  perBiasThres = 20,
  considerPerBiasCV = TRUE,
  perBiasDistThres = 10,
  RfThresL = 80,
  RfThresU = 120
)

Arguments

D

data.frame data set, e.g. result of readDataTable or readDataSE. Has to include exactly two columns: "Concentration" and "Measurement".
substance

character(1)
Name of the substance (default is "substance"). Will be added to the result files and may be used when plotting multiple calibration curves in one plot.
minReplicates

integer(1)
Minimal number of replicates per concentration level. Concentration levels with too few data points will be removed.
cvThres

numeric(1)
Threshold for CV per concentration level in percent (default is 20).
calcContinuousPrelimRanges

logical(1)
If TRUE, the longest continuous range is selected (default is TRUE). If FALSE, gaps with CVs larger than the threshold may be included.
weightingMethod

character(1)
Method for weighting (currently "1/x", "1/x^2" and "None" are supported, default is 1/x^2).
centralTendencyMeasure

character(1)
Method for calculating average percent bias, "mean" (default) or "median".
perBiasThres

numeric(1)
Threshold for average percent bias in percent, default is 20.
considerPerBiasCV

logical(1)
If TRUE, CV is considered for the elimination of the concentration level (default). CV will only be considered if the difference in percent bias values is lower than perBiasDistThres.
perBiasDistThres

numeric(1)
Threshold for the difference in average percent bias in percent (for lower differences, CV will be considered), default is 10.
RfThresL

numeric(1)
Lower threshold for response factor in percent (default is 80).
RfThresU

numeric(1)
Upper threshold for response factor in percent (default is 120).

Value

List with the following elements:

  • mod: lm-object containing the final linear model.

  • final_linear_range: vector of concentration levels that are part of the final linear range.

  • weightingMethod: weighting method used for the linear model.

  • result_table_conc_levels: Result table with one line for each concentration level (generated by assemble_results).

  • result_table_obs: Result table with one line per observation (e.g. individual response factors for each data point) (generated by assemble_results).

Examples

data_path <- system.file("extdata", "MSQC1_xlsx", "GGPFSDSYR_QTRAP_y5.xlsx",
        package = "CalibraCurve")
D <- readDataTable(dataPath = data_path, concCol = 16, measCol = 12,
        fileType = "xlsx")
calc_single_curve(D = D)

Calculate Response factors

Description

Final linear range: Function, which returns a list with response factor values for a data set (given as list)

Usage

calcRF(x, mod)

Arguments

x

list of data.frames
List of data.frames containing data for each concentration level (result from cleanData).
mod

lm object
Final linear model fit (object "mod" from results of calculate_FLR).

Value

List with the following elements:

  • RFs: List with response factors for each concentration level.

  • meanRFs: Vector with mean response factors for each concentration level.

response factor values for each concentration level.

Examples

file <- system.file("extdata", "MSQC1", "msqc1_dil_GGPFSDSYR.rds",
package = "CalibraCurve")
D_list <- readDataSE(file, concColName = "amount_fmol",
        substColName = "Substance", assayNumber = 1)
data_cleaned <- cleanData(D_list[[1]])
RES_PLR <- calculate_PLR(data_cleaned, calcContinuousPrelimRanges = FALSE)
RES_FLR <- calculate_FLR(RES_PLR$dataPrelim)

calcRF(data_cleaned, mod = RES_FLR$mod)

Calculate the final linear range

Description

Calculate the final linear range

Usage

calculate_FLR(
  dataPrelim,
  weightingMethod = "1/x^2",
  centralTendencyMeasure = "mean",
  perBiasThres = 20,
  considerPerBiasCV = TRUE,
  perBiasDistThres = 10
)

Arguments

dataPrelim

list of data.frames
List of data.frames containing data only within the preliminary linear range (result from calculate_PLR).
weightingMethod

character(1)
Method for weighting (currently "1/x", "1/x^2" and "None" are supported, default is 1/x^2).
centralTendencyMeasure

character(1)
Method for calculating average percent bias, "mean" (default) or "median".
perBiasThres

numeric(1)
Threshold for average percent bias in percent, default is 20.
considerPerBiasCV

logical(1)
If TRUE, CV is considered for the elimination of the concentration level (default). CV will only be considered if the difference in percent bias values is lower than perBiasDistThres.
perBiasDistThres

numeric(1)
Threshold for the difference in average percent bias in percent (for lower differences, CV will be considered), default is 10. Only relevant if considerPerBiasCV = TRUE.

Value

List with the following elements:

  • dataFinal: List of data.frames containing data only within the final linear range.

  • mod: lm-object containing the final linear model (weighted or unweighted, depending on weightingMethod).

  • perBias: result list of .calcPerBiasLevels.

  • perBiasAvgSDCV: result data.frame of .calcPerBiasAvgSDCV.

Examples

file <- system.file("extdata", "MSQC1", "msqc1_dil_GGPFSDSYR.rds",
        package = "CalibraCurve")
D_list <- readDataSE(file, concColName = "amount_fmol",
        substColName = "Substance", assayNumber = 1)
data_cleaned <- cleanData(D_list[[1]])
RES_PLR <- calculate_PLR(data_cleaned, calcContinuousPrelimRanges = FALSE)

calculate_FLR(RES_PLR$dataPrelim)

Calculate preliminary linear range (PLR)

Description

Calculate preliminary linear range (PLR)

Usage

calculate_PLR(dataCleaned, cvThres = 20, calcContinuousPrelimRanges = TRUE)

Arguments

dataCleaned

data.frame
Data cleaned by cleanData.
cvThres

numeric(1)
Threshold for CV per concentration level in percent (default is 20).
calcContinuousPrelimRanges

logical(1)
If TRUE, the longest continuous range fulfilling the CV threshold is selected (default is TRUE). If FALSE, gaps with CVs larger than the threshold may be included.

Value

List with the following elements:

  • dataPrelim: List of data.frames containing data only within the preliminary linear range.

  • concLevelsCV: Vector with the calculated CV for each concentration level.

  • prelimConcLevels: Vector with the concentration levels within the preliminary linear range.

Examples

file <- system.file("extdata", "MSQC1", "msqc1_dil_GGPFSDSYR.rds",
package = "CalibraCurve")
D_list <- readDataSE(file, concColName = "amount_fmol",
        substColName = "Substance", assayNumber = 1)
data_cleaned <- cleanData(D_list[[1]])

calculate_PLR(data_cleaned, calcContinuousPrelimRanges = FALSE)

CalibraCurve

Description

CalibraCurve

Usage

CalibraCurve(
  D_list,
  output_path = NULL,
  substance = "substance",
  minReplicates = 3,
  cvThres = 20,
  calcContinuousPrelimRanges = FALSE,
  weightingMethod = "1/x^2",
  centralTendencyMeasure = "mean",
  perBiasThres = 20,
  considerPerBiasCV = TRUE,
  perBiasDistThres = 10,
  RfThresL = 80,
  RfThresU = 120,
  plot_type = "single_plots",
  RF_colour_threshold = "orange",
  RF_colour_within = "#00BFC4",
  RF_colour_outside = "#F8766D",
  device = "png",
  CC_plot_width = 15,
  CC_plot_height = 10,
  RF_plot_width = 15,
  RF_plot_height = 10,
  base_size = 11,
  RF_legend = FALSE,
  plot_dpi = 300,
  verbose = TRUE,
  ...
)

Arguments

D_list

data.frame/list
Data.frame or list of data.frames (one for one specific substance, each must contain two columns: Concentration and Measurement)
output_path

character(1)
Folder to save results (table and plots). If NULL (default), results are not saved.
substance

character(1)
Name of the substance (default is "substance1"). Will be added to the result files and will be used as labels in the plots.
minReplicates

integer(1)
Minimal number of replicates/data points per concentration level. Concentration levels with too few data points will be removed.
cvThres

numeric(1)
Threshold for CV per concentration level in percent (default is 20).
calcContinuousPrelimRanges

logical(1)
If TRUE, the longest continuous range is selected (default is TRUE). If FALSE, gaps with CVs larger than the threshold may be included.
weightingMethod

character(1)
Method for weighting (currently "1/x", "1/x^2" and "None" are supported, default is 1/x^2).
centralTendencyMeasure

character(1)
Method for calculating average percent bias, "mean" (default) or "median".
perBiasThres

numeric(1)
Threshold for average percent bias in percent, default is 20.
considerPerBiasCV

logical(1)
If TRUE, CV is considered for the elimination of the concentration level (default). CV will only be considered if the difference in percent bias values is lower than perBiasDistThres.
perBiasDistThres

numeric(1)
Threshold for the difference in average percent bias in percent (for lower differences, CV will be considered), default is 10.
RfThresL

numeric(1)
Lower threshold for response factor in percent (default is 80).
RfThresU

numeric(1)
Upper threshold for response factor in percent (default is 120).
plot_type

character(1)
Type of plot for calibration curves: "single_plots" (default, generate a separate plot for each substance), "multiplot" (generate a graphic with subplots for each substance) or "all_in_one" (generate a single plot with all substances).
RF_colour_threshold

character(1)
Response factor plot: Colour for horizontal threshold lines, default is "orange".
RF_colour_within

character(1)
Response factor plot: Colour for points and lines within the final linear range, default is "#00BFC4" (default ggplot colour).
RF_colour_outside

character(1)
Response factor plot: Colour for horizontal outside of the final linear range, default is "#F8766D" (default ggplot colour).
device

character(1)
Device for saving the plot (default is "png"). Other options include "pdf", "jpeg", "tiff", "svg" etc. For details see ggsave.
CC_plot_width

numeric(1)
Plot width in cm (default is 10).
CC_plot_height

numeric(1)
Plot height in cm (default is 10).
RF_plot_width

numeric(1)
Plot width in cm (default is 10).
RF_plot_height

numeric(1)
Plot height in cm (default is 10).
base_size

numeric(1)
Base size for the plot theme, default is 11.
RF_legend

logical(1)
If TRUE, a legend will be added to the response factor plots.
plot_dpi

numeric(1)
Plot resolution in dpi (default is 300).
verbose

logical(1)
If FALSE, no messages will be printed.
...

additional parameters for plotCalibraCurve

Value

List with the following elements:

  • RES: List of CalibraCurve results (one item per substance, output from calc_single_curve).

  • summary_tab: Data.frame with summary information about the calibration curves, one row per substance.

  • plot_CC_list: List of ggplot2 objects of the calibration curves (one per substance for "single_plots", otherwise only one item).

  • plot_RF_list: List of ggplot2 objects of the response factor plots (one item).

Examples

### NOTE: as output_path is not specified here, no files will be saved.
### Set output_path to a folder of your choice to save the results.

### single xlsx file:
data_path <- system.file("extdata", "MSQC1_xlsx", "GGPFSDSYR_QTRAP_y5.xlsx",
    package = "CalibraCurve")
D <- readDataTable(dataPath = data_path, concCol = 16, measCol = 12,
    fileType = "xlsx")
RES <- CalibraCurve(D_list = D)
RES$plot_CC_list
RES$plot_RF_list

### multiple xlsx files (in a folder) as multiplot:
data_folder <- system.file("extdata", "MSQC1_xlsx", package = "CalibraCurve")
D_list <- readMultipleTables(dataFolder = data_folder, fileType = "xlsx",
    concCol = 16, measCol = 12)
RES <- CalibraCurve(D_list = D_list, plot_type = "multiplot")
RES$plot_CC_list

### single rds file (SummarizedExperiment) as an all-in-one plot:
file <- system.file("extdata", "MSQC1", "msqc1_dil_GGPFSDSYR.rds",
    package = "CalibraCurve")
D_list <- readDataSE(dataPath = file, concColName = "amount_fmol",
        substColName = "Substance", assayNumber = 1)
RES <- CalibraCurve(D_list, plot_type = "all_in_one")
RES$plot_CC_list

Clean data (remove 0s and NAs, remove concentration levels with insufficient number of replicates)

Description

Clean data (remove 0s and NAs, remove concentration levels with insufficient number of replicates)

Usage

cleanData(rawData, minReplicates = 3)

Arguments

rawData

data.frame
data set to be cleaned, result of readDataTable or readDataSE.
minReplicates

integer(1)
Minimal number of replicates per concentration level. Concentration levels with too few data points will be removed.

Value

list of data.frames, each element contains data for a specific concentration level

Examples

file <- system.file("extdata", "MSQC1", "msqc1_dil_GGPFSDSYR.rds",
    package = "CalibraCurve"
)
D_list <- readDataSE(file,
    concColName = "amount_fmol",
    substColName = "Substance", assayNumber = 1
)
cleanData(D_list[[1]])

Plot the calibration curve

Description

Plot the calibration curve

Usage

plotCalibraCurve(
  CC_RES,
  ylab = "Intensity",
  xlab = "Concentration",
  show_regression_info = FALSE,
  show_linear_range = TRUE,
  show_data_points = TRUE,
  plot_type = "multiplot",
  point_colour = "black",
  curve_colour = "red",
  linear_range_colour = "black",
  multiplot_nrow = NULL,
  multiplot_ncol = NULL,
  multiplot_scales = "free",
  regression_info_size = 3,
  base_size = 11
)

Arguments

CC_RES

list
Results of CalibraCurve. Each list element is the result of calc_single_curve
ylab

character(1)
y-axis label.
xlab

character(1)
x-axis label.
show_regression_info

logical(1)
If TRUE, show regression information (R2, slope, intercept) on the plot.
show_linear_range

logical(1)
If TRUE, show the linear range of the calibration curve as a rectangle in the plot.
show_data_points

logical(1)
If TRUE, show the data points on the plot.
plot_type

character(1)
Type of plot for calibration curves: "single_plots" (default, generate a separate plot for each substance), "multiplot" (generate a graphic with subplots for each substance) or "all_in_one" (generate a single plot with all substances).
point_colour

character(1)
Colour of the data points, default is "black".
curve_colour

character(1)
Colour of the calibration curve, default is "red".
linear_range_colour

character(1)
Colour of the linear range background, default is "black" (colour is weakened by alpha = 0.1).
multiplot_nrow

integer(1)
Number of rows for the multiplot layout (default is NULL, which means that there is only one row).
multiplot_ncol

integer(1)
Number of columns for the multiplot layout (default is NULL, which means that the number of columns is determined automatically based on the number of curves).
multiplot_scales

character(1)
Scales for the multiplot layout, default is "free" (which means that each plot has its own scales). Other options are "fixed", "free_x", "free_y".
regression_info_size

numeric(1)
Size of the regression information text on the plot, default is 3.
base_size

numeric(1)
Base size for the plot theme, default is 11.

Value

List of a ggplot2 object containing the calibration curve plot (CC_plot) and a data.frame containing summary data for annotations (annotation_dat).

Examples

file <- system.file("extdata", "MSQC1", "msqc1_dil_GGPFSDSYR.rds",
    package = "CalibraCurve"
)
D_list <- readDataSE(file,
    concColName = "amount_fmol",
    substColName = "Substance", assayNumber = 1
)
CC_RES <- calc_single_curve(D_list[[1]], calcContinuousPrelimRanges = FALSE)

plotCalibraCurve(list(substance = CC_RES))

Plot response factors

Description

Plot response factors

Usage

plotResponseFactors(
  RES,
  RfThresL = 80,
  RfThresU = 120,
  ylab = "Response Factor",
  xlab = "Concentration",
  colour_threshold = "orange",
  colour_within = "#00BFC4",
  colour_outside = "#F8766D",
  legend = FALSE,
  base_size = 11,
  substance = NULL
)

Arguments

RES

list
Results of calc_single_curve.
RfThresL

numeric(1)
Lower threshold for response factor in percent (default is 80).
RfThresU

numeric(1)
Upper threshold for response factor in percent (default is 120).
ylab

character(1)
y-axis label.
xlab

character(1)
x-axis label.
colour_threshold

character(1)
Colour for horizontal threshold lines, default is "orange".
colour_within

character(1)
Colour for points and lines within the threshold, default is "#00BFC4" (default ggplot colour).
colour_outside

character(1)
Colour for horizontal outside of the threshold, default is "#F8766D" (default ggplot colour).
legend

logical(1)
If TRUE, a legend is added to the plot, default is FALSE.
base_size

numeric(1)
Base size for the plot theme, default is 11.
substance

character(1)
Name of the substance (default is "substance1"). Will be added to the result files and will be used as labels in the plots.

Value

A ggplot2 object containing the response factor plot.

Examples

file <- system.file("extdata", "MSQC1", "msqc1_dil_GGPFSDSYR.rds",
    package = "CalibraCurve"
)
D_list <- readDataSE(file,
    concColName = "amount_fmol",
    substColName = "Substance", assayNumber = 1
)
CC_RES <- calc_single_curve(D_list[[1]], calcContinuousPrelimRanges = FALSE)

plotResponseFactors(RES = CC_RES)

Predict concentrations from intensity values based on a given calibration curve

Description

Predict concentrations from intensity values based on a given calibration curve

Usage

predictConcentration(CC_res, newdata, verbose = TRUE)

Arguments

CC_res

list
Results of CalibraCurve.
newdata

numeric
A vector of intensity values for which to predict concentrations.
verbose

logical(1)
If TRUE, a warning message is given if estimated concentrations are outside of the linear range.

Details

The function will give a warning if any of the predicted concentrations are outside the final linear range. This is important to ensure that the predictions are reliable and within the linear range of the calibration curve.

Value

A data frame with the following columns:

  • intensity: The input intensity values.

  • predicted_concentrations: The predicted concentrations based on the calibration curve.

  • linear_range: A logical vector indicating whether the predicted concentrations are within the final linear range.

Examples

file <- system.file("extdata", "MSQC1", "msqc1_dil_GGPFSDSYR.rds",
        package = "CalibraCurve")
D <- readDataSE(file, concColName = "amount_fmol",
        substColName = "Substance")
RES <- CalibraCurve(D)
newdata <- c(1000000, 10000000, 100000000) # 1e6, 1e7, 1e8
predictConcentration(RES$RES[[4]], newdata = newdata)

Read data stored as an SummarizedExperiment object (directly or stored in an .rds file). Extracts the two relevant columns (concentration and measurement) and orders the data by increasing concentration level.

Description

Read data stored as an SummarizedExperiment object (directly or stored in an .rds file). Extracts the two relevant columns (concentration and measurement) and orders the data by increasing concentration level.

Usage

readDataSE(
  dataPath = NULL,
  rawDataSE = NULL,
  concColName,
  substColName,
  assayNumber = 1,
  rowNumbers = NULL
)

Arguments

dataPath

character(1)
Path to the data file (.rds file)
rawDataSE

SummarizedExperiment
SummarizedExperiment object
concColName

character(1)
Name of the column in the colData() containing the concentration levels.
substColName

character(1)
column name of rowData() containing the substance name (must be a unique value in each row)
assayNumber

integer(1)
Number of assay to be extracted from the SummarizedExperiment object
rowNumbers

integer
Row numbers to extract from the SummarizedExperiment object. Default is NULL, which means that all rows in the object will be used.

Details

The SummarizedExperiments object may contain quantitative values from targeted proteomics, lipidomics or metabolomics experiments. The colData has to contain a column with the concentration levels (concColName). The rowData has to contain a column with the substance names (e.g. peptide sequence, name of lipid or metabolite etc).

Value

List of data.frame, each with two numeric columns: Concentration and Measurement

Examples

file <- system.file("extdata", "MSQC1", "msqc1_dil_GGPFSDSYR.rds",
    package = "CalibraCurve")

D_list <- readDataSE(file,
    concColName = "amount_fmol",
    substColName = "Substance", assayNumber = 1)

# Alternative: import SummarizedExperiment object directly
rawDataSE <- readRDS(file)

D_list2 <- readDataSE(rawDataSE = rawDataSE,
    concColName = "amount_fmol",
    substColName = "Substance", assayNumber = 1
)

Read data in different table input formats (xlsx, csv or txt). Extracts the two relevant columns (concentration and measurement) and orders the data by increasing concentration.

Description

Read data in different table input formats (xlsx, csv or txt). Extracts the two relevant columns (concentration and measurement) and orders the data by increasing concentration.

Usage

readDataTable(
  dataPath,
  fileType,
  concCol,
  measCol,
  substCol = NULL,
  sep = ",",
  dec = ".",
  header = TRUE,
  naStrings = c("NA", "NaN", "Filtered", "#NV"),
  sheet = 1
)

Arguments

dataPath

character(1)
Path to the data file (.csv, .txt or .xlsx file).
fileType

character(1)
Type of file: "csv", "txt" or "xlsx".
concCol

integer(1)
Column number of the concentration values.
measCol

integer
Column number of the concentration values.
substCol

integer
Column number of the substance names (optional). If not set, all rows will be interpreted as belonging to the same substance.
sep

character(1)
The field separator, default is ",".
dec

character(1)
Decimal separator, default is ".".
header

logical(1)
If TRUE, first line is counted as column names. The default is TRUE.
naStrings

character
Vector of strings which are to be interpreted as NA. The default is c("NA", "NaN", "Filtered", "#NV").
sheet

integer(1)
Sheet number (only needed for xlsx files, default is to use the first sheet).

Value

Data.frame with two numeric columns: Concentration and Measurement

Examples

### xlsx file:
file <- system.file("extdata", "MSQC1_xlsx", "GGPFSDSYR_QTRAP_y5.xlsx",
    package = "CalibraCurve"
)
D <- readDataTable(file, fileType = "xlsx", concCol = 16, measCol = 12)

Read folder of files in different table input formats (xlsx, csv or txt).

Description

Read folder of files in different table input formats (xlsx, csv or txt).

Usage

readMultipleTables(dataFolder, fileType, concCol, measCol, ...)

Arguments

dataFolder

character(1)
Folder containing either xlsx, csv or txt files
fileType

character(1)
Type of file: "csv", "txt" or "xlsx".
concCol

integer(1)
Column number of the concentration values.
measCol

integer
Column number of the concentration values.
...

additional parameters to readDataTable

Value

List of data.frame, each with two numeric columns: Concentration and Measurement

Examples

data_folder <- system.file("extdata", "MSQC1_xlsx",
    package = "CalibraCurve")
D_list <- readMultipleTables(
    dataFolder = data_folder, fileType = "xlsx",
    concCol = 16, measCol = 12
)