--- title: "MQmetrics" author: "Alvaro Sanchez-Villalba" output: BiocStyle::html_document vignette: > %\VignetteIndexEntry{MQmetrics} %\VignetteEngine{knitr::rmarkdown} %\VignetteEncoding{UTF-8} --- ```{r, include = FALSE} knitr::opts_chunk$set( collapse = TRUE, warning = FALSE, comment = "#>" ) ``` # Overview The package MQmetrics (MaxQuant metrics) provides a workflow to analyze the quality and reproducibility of your proteomics mass spectrometry analysis from MaxQuant. Input data are extracted from several [MaxQuant](https://pubmed.ncbi.nlm.nih.gov/27809316/) output tables ([MaxQUant Summer School Output tables](https://www.youtube.com/watch?v=wEIWRFKlzJ0)), and produces a pdf report. It includes several visualization tools to check numerous parameters regarding the quality of the runs. It also includes two functions to visualize the [iRT peptides from Biognosys](https://biognosys.com/product/irt-kit/) in case they were spiked in the samples. # Workflow ## Install the package You can install MQmetrics from Biocodunctor with: ```r if (!requireNamespace("BiocManager", quietly = TRUE)) install.packages("BiocManager") BiocManager::install("MQmetrics") ``` You can install the development version from [GitHub](https://github.com/svalvaro/) with: ``` {r install, eval = FALSE} # install.packages("devtools") devtools::install_github("svalvaro/MQmetrics") ``` # Data Input After your MaxQuant run has finished, a folder named **combined** has been created. This folder should have at least two other folders within: ../combined/txt/ Containing all the tables.txt ../combined/proc/ Containing #runningTimes.txt You just need the path to the **combined** folder and you will be able to start using MQmetrics. ```{r MQPathCombined} MQPathCombined <- '/home/alvaro/Documents/MaxQuant/example5/combined/' ``` ## Generate a report First you need to load the library. ```{r report, message=FALSE} library(MQmetrics) ``` Then you just need to use the `generateReport()` function. This function has parameters to control each of the function that it aggregates. You can read more about those parameters by using: ```{r report_question_mark, eval=FALSE} ?generateReport ``` Though, the most important parameters are the following: ```{r parameters, eval=FALSE} generateReport(MQPathCombined = , # directory to the combined folder output_dir = , # directory to store the resulting pdf long_names = , # If your samples have long names set it to TRUE sep_names = , # Indicate the separator of those long names UniprotID = , # Introduce the UniprotID of a protein of interest intensity_type = ,) # Intensity or LFQ # The only mandatory parameter is MQPathCombined, the rest are optional. ``` Is as simple as this to use MQmetrics: ```{r example,eval=FALSE} # If you're using a Unix-like OS use forward slashes. MQPathCombined <- '/home/alvaro/Documents/MaxQuant/example5/combined/' # If you're using Windows you can also use forward slashes: MQPathCombined <- "D:/Documents/MaxQuant_results/example5/combined/" #Use the generateReport function. generateReport(MQPathCombined) ``` ## Visualization plots If you are only interested in a few plots from the `generateReport()` function, you can do it. You only need to have access to each file independently. ### Load the data MQmetrics requires 8 tables from the MaxQuant analysis and the #runningTimes file. If you want to learn more about the information of each of these tables, you can do so in the [MaxQuant Summer School videos](https://www.youtube.com/watch?v=wEIWRFKlzJ0). ```{r, message=FALSE, warning=FALSE} # To create the vignettes and examples I use data that is in the package itself: MQPathCombined <- system.file('extdata/combined/', package = 'MQmetrics') # make_MQCombined will read the tables needed for creating the outputs. MQCombined <- make_MQCombined(MQPathCombined, remove_contaminants = TRUE) ``` ### MaxQuant Analysis Parameters ```{r ExperimentDuration, comment= NA} MaxQuantAnalysisInfo(MQCombined) ``` # Visualizations. ## Proteins Identified The function `PlotProteinsIdentified()`, will take as input the proteinGroups.txt table and show the number of proteins and NAs in each sample. It can differentiate two types of intensities: 'Intensity' or 'LFQ'. ```{r PlotProteins,fig.height=5, fig.width=7.2} PlotProteinsIdentified(MQCombined, long_names = TRUE, sep_names = '_', intensity_type = 'LFQ', palette = 'Set2') ``` ## Peptides Identified The function `PlotPeptidesIdentified()`, will take as input the summary table and show the number of peptides sequences identified in each sample. ```{r PeptidesIdentified, warning=FALSE , fig.height=5, fig.width=7.2} PlotPeptidesIdentified(MQCombined, long_names = TRUE, sep_names = '_', palette = 'Set3') ``` ## Proteins versus peptide/protein ratio The function `PlotIdentificationRatio()`, will take as input the summary and proteinGroups tables and plot the number of protein found vs the ratio of peptides/proteins found in each Experiment. ```{r ratio, warning=FALSE , fig.height=5, fig.width=7.2} PlotProteinPeptideRatio(MQCombined, intensity_type = 'LFQ', long_names = TRUE, sep_names = '_') ``` ## MS/MS submitted versus identified The function `PlotMsMs()`, will take as input the summary.txt table and show the number of MS/MS Submitted and identified in each sample. ```{r PlotMSMS, fig.height=5, fig.width=7.2} PlotMsMs(MQCombined, long_names = TRUE, sep_names = '_', position_dodge_width = 1, palette = 'Set2') ``` ## Peaks submitted versus identified The function `PlotPeaks()`, will take as input the summary.txt table and show the number peaks detected and sequenced in each sample. ```{r PlotPeaks, fig.height=5, fig.width=7.2} PlotPeaks(MQCombined, long_names = TRUE, sep_names = '_', palette = 'Set2') ``` ## Isotope patterns detected and sequenced The function `PlotIsotopePattern()`,will take as input the summary.txt table and show the number isotope patterns detected and sequenced in each sample. ```{r isotope, fig.height=5, fig.width=7.2} PlotIsotopePattern(MQCombined, long_names = TRUE, sep_names = '_', palette = 'Set2') ``` ## Charge-state of the precursor ions The function `PlotCharge()`, will take as input the evidence.txt table and show the charge-state of the precursor ion in each sample. ```{r Charg, warning=FALSE, message=FALSE, fig.height=10, fig.width=7.2} PlotCharge(MQCombined, palette = 'Set2', plots_per_page = 6) ``` ## Protease Specificity The function `PlotProteaseSpecificity()`, will take as input the summary.txt table and show the number peaks detected and sequenced in each sample. ```{r missed_cleavages,fig.height=10, fig.width=7.2} PlotProteaseSpecificity(MQCombined, palette = 'Set2', plots_per_page = 6) ``` ## Peptide Hydrophobicity The function `PlotHydrophobicity()`, takes as input the peptides.txt table and returns the distribution of GRAVY score. ```{r PlotHydrophobicity, warning= F, message= F,fig.height=10, fig.width=7.2} PlotHydrophobicity(MQCombined, show_median = TRUE, binwidth = 0.1, size_median = 1.5, palette = 'Set2', plots_per_page = 6) ``` ## Andromeda Score The function `PlotAndromedaScore()`, takes as input the peptides.txt table and returns the distribution of MaxQuant's Andromeda Score. ```{r AndromedaScore, message = F, warning = FALSE,fig.height=10, fig.width=7.2} PlotAndromedaScore(MQCombined, palette = 'Set2', plots_per_page = 6) ``` ## Protein intensities comparison The function `PlotIntensity()`, takes as input the proteinGroups.txt table and returns a violin plot for those intensities. If the 'LFQ' intensities are in the proteinGroups.txt table, it will by default split the violion into "LFQ' and 'Intensity'. The parameter split_violin_intensity, can be set to FALSE and then select wether you would like to visualize the 'Intensity' or 'LFQ' intensity individually. If split_violin_intensity is set TRUE, but no LFQ intensities are not present, it will automatically show the normal Intensities. ```{r PlotIntensity, warning = FALSE, fig.height=5, fig.width=7.2} PlotIntensity(MQCombined, split_violin_intensity = TRUE, intensity_type = 'LFQ', log_base = 2, long_names = TRUE, sep_names = '_', palette = 'Set2') ``` ## Protein PCA The function `PlotPCA()` takes as input the proteinGroups.txt table and creates a Principal Componente Analysis plot of each Experiment. ```{r PlotPCA, warning = FALSE, fig.height=5, fig.width=7.2} PlotPCA(MQCombined, intensity_type = 'LFQ', palette = 'Set2') ``` ## Dynamic Range (Combined) The function `PlotCombinedDynamicRange()` takes as input the proteinGroups.txt table and returns the dynamic range of all experiments combined. If the parameter show_shade is used, a square will appear showing the percent_proteins selected and the orders of abundance. ```{r DynamicRange, fig.height=5, fig.width=7.2} PlotCombinedDynamicRange(MQCombined, show_shade = TRUE, percent_proteins = 0.90) ``` ## Dynamic Range (Individual) The function `PlotAllDynamicRange()` takes as input the proteinGroups.txt table and returns the dynamic range of all experiments separated. If the parameter show_shade is used, a square will appear showing the *percent_proteins* selected and the orders of abundance. ```{r DynamicRangeAll, fig.height=10, fig.width=7.2} PlotAllDynamicRange(MQCombined, show_shade = TRUE, percent_proteins = 0.90) ``` ## Protein Overlap The function `PlotProteinOverlap()` takes as input the proteinGroups.txt table and returns a plot that shows the number of proteins identified in the samples. ```{r protein_coverage_all, fig.height=7.2, fig.width=7.2} PlotProteinOverlap(MQCombined) ``` ## Protein Coverage The function `PlotProteinCoverage()` takes as input the peptides.txt and proteinGroups.txt tables and one or multiple protein UniprotID(s). It shows, if present, the coverage of that protein in each of the samples. ```{r protein_degradation,fig.height=12, fig.width=12, message=FALSE, warning=F} PlotProteinCoverage(MQCombined, UniprotID = c('P55072', 'P13639'), log_base = 2, segment_width = 1, palette = 'Set2', plots_per_page = 6) ``` ## iRT peptides (1) The function `PlotiRT()` takes as input the evidence.txt table and returns, if found the iRT peptides from Biognosys. Their retention time and intensity. ```{r irt_peps1, warning = FALSE, message= FALSE, fig.height=10, fig.width=7.2} PlotiRT(MQCombined, show_calibrated_rt = FALSE, tolerance = 0.001, plots_per_page = 6) ``` ## iRT peptides (2) The function `PlotiRTScore()` takes as input the evidence.txt table and returns, if found, a linear regression of the retention times of the iRT peptides of Biognosys. ```{r irt_peps2, warning = FALSE, message= FALSE, fig.height=10, fig.width=7.2} PlotiRTScore(MQCombined, tolerance = 0.001, plots_per_page = 6) ``` ## Total Ion Current The function `PlotTotalIonCurrent()` takes as input the msmsScans.txt, and returns a plot showing the TIC values vs the retention time of each sample. It can show as well the maximum value of each sample. ```{r TotalIonCurrent, fig.height=10, fig.width=7.2} PlotTotalIonCurrent(MQCombined, show_max_value = TRUE, palette = 'Set2', plots_per_page = 6) ``` ## Acquisition Cycle The function `PlotAcquisitionCycle` takes as input the msScans.txt table and returns the cycle time and MS/MS count vs the retention time of each sample. **Note**: msScans.ttxt is not generated automatically by MaxQuant, the user must select it in MaxQuant: Global Parameters --> tables ```{r Pn, fig.height= 10, warning=F, message=F,fig.height=10, fig.width=7.2} PlotAcquisitionCycle(MQCombined, palette = 'Set2', plots_per_page = 6) ``` ## Post-translational modifications The function `PlotPTM()`, takes as input the modificationSpecificPeptides.txt table and returns the main modifications found at the peptide level. The parameters can be adjusted to select the minimum number of peptides modified per group, and whether or not you would like to visualize the Unmodified peptides. ```{r PTM, fig.height=8, fig.width=7.2} PlotPTM(MQCombined, peptides_modified = 1, plot_unmodified_peptides = FALSE, palette = 'Set2', aggregate_PTMs = TRUE, plots_per_page = 6) ``` ## Post-translational modifications across samples The function `PlotPTMAcrossSamples()`, takes as input the modificationSpecificPeptides.txt table and one or multiple modification of interest. It returns a plot showing the intensities of that given PTM across the samples. It is important to specify correctly the name of the PTM, the same way as MaxQuant: ```{r ptm_single, fig.height=8, fig.width=7.2} PlotPTMAcrossSamples(MQCombined, PTM_of_interest = c('Acetyl (Protein N-term)', 'Oxidation (M)'), long_names = TRUE, sep_names = '_') ``` The parameters can be adjusted to select the minimum number of peptides modified per group, and whether or not you would like to visualize the Unmodified peptides. # Helper Functions This package provides two extra functions to helps to analyze the proteomics data from MaxQuant: The function `make_MQCombined()` takes as input the path to the *combined* folder resulting from MaxQuant analysis. It will read the tables needed and by default remove the potential contaminants, Reverse, and proteins identified only by site. The function `ReportTables()` takes as input the path to the **combined** folder, and returns tables with information needed to create some of the most important plots in this package. ```{r report_tables, message=FALSE} ReportTables(MQCombined, log_base = 2, intensity_type = 'Intensity') ``` # Session Info ```{r add_session_info} sessionInfo() ```