mpwR

R-CMD-check CRAN status

mpwR [ɪmˈpaʊə(r)] offers a systematic approach for comparing proteomic workflows and empowers the researcher to effortlessly access valuable information about identifications, data completeness, quantitative precision, and other performance indicators across an unlimited number of analyses and multiple software tools. It can be used to analyze label-free mass spectrometry-based experiments with data-dependent or data-independent spectral acquisition.

Applications

The functions of mpwR provide a great foundation to generate customized reports e.g. with RMarkdown or to build shiny apps/dashboards for downstream data analysis. An example for a shiny dashboard is available here.

Installation

Install the development version from GitHub using the devtools package by using the following commands:

# install.packages("devtools") #remove "#" if you do not have devtools package installed yet
devtools::install_github("OKdll/mpwR", dependencies = TRUE) # use dependencies TRUE to install all required packages for mpwR

Preparation

Requirements

As input the standard outputs of ProteomeDiscoverer, Spectronaut, DIA-NN or MaxQuant are supported by mpwR. Details about further requirements are listed in the vignette Requirements.

Import

Importing the output files from each software can be easily performed with prepare_mpwR. Further details about importing data are highlighted in the vignette Import.

files <- prepare_mpwR(path = "Path_to_Folder_with_files")

Load packages

library(mpwR)
library(flowTraceR)
library(magrittr) 
library(dplyr)
library(tidyr)
library(stringr)
library(tibble)
library(ggplot2)

Example - Workflow

This is a basic example which shows the downstream analysis for number of identifications and data completeness. Please check the vignette Workflow for a detailed analysis pipeline and more functionalities.

#get example
files <- create_example()

Number of Identifications

Report

The number of identifications can be determined with get_ID_Report.

ID_Reports <- get_ID_Report(input_list = files)

 

For each analysis an ID Report is generated and stored in a list. Each ID Report entry can be easily accessed:

ID_Reports[["DIA-NN"]]
#>   Analysis Run ProteinGroup.IDs Protein.IDs Peptide.IDs Precursor.IDs
#> 1   DIA-NN R01                5           5           5             5
#> 2   DIA-NN R02                5           5           5             5

 

Plot

Individual

Each ID Report can be plotted with plot_ID_barplot from precursor- to proteingroup-level. The generated barplots are stored in a list.

ID_Barplots <- plot_ID_barplot(input_list = ID_Reports, level = "ProteinGroup.IDs")

 

The individual barplots can be easily accessed:

ID_Barplots[["DIA-NN"]]

 

Summary

As a visual summary a boxplot can be generated with plot_ID_boxplot.

plot_ID_boxplot(input_list = ID_Reports, level = "ProteinGroup.IDs")

 

Data Completeness

Report

Data Completeness can be determined with get_DC_Report for absolute numbers or in percentage.

DC_Reports <- get_DC_Report(input_list = files, metric = "absolute")
DC_Reports_perc <- get_DC_Report(input_list = files, metric = "percentage")

 

For each analysis a DC Report is generated and stored in a list. Each DC Report entry can be easily accessed:

DC_Reports[["DIA-NN"]]
#>   Analysis Nr.Missing.Values Precursor.IDs Peptide.IDs Protein.IDs
#> 1   DIA-NN                 1             0           0           4
#> 2   DIA-NN                 0             5           5           3
#>   ProteinGroup.IDs  Profile
#> 1                2   unique
#> 2                4 complete

 

Plot

Individual

Absolute

Each DC Report can be plotted with plot_DC_barplot from precursor- to proteingroup-level. The generated barplots are stored in a list.

DC_Barplots <- plot_DC_barplot(input_list = DC_Reports, level = "ProteinGroup.IDs", label = "absolute")

 

The individual barplots can be easily accessed:

DC_Barplots[["DIA-NN"]]

 

Percentage

plot_DC_barplot(input_list = DC_Reports_perc, level = "ProteinGroup.IDs", label = "percentage")[["DIA-NN"]]

 

Summary

As a visual summary a stacked barplot can be generated with plot_DC_stacked_barplot.

Absolute

plot_DC_stacked_barplot(input_list = DC_Reports, level = "ProteinGroup.IDs", label = "absolute")

 

Percentage

plot_DC_stacked_barplot(input_list = DC_Reports_perc, level = "ProteinGroup.IDs", label = "percentage")