---
title: "MsQuality: Calculation of QC metrics from mass spectrometry data"
author:
 - name: Thomas Naake and Johannes Rainer
   mail: thomas.naake@embl.de, thomasnaake@googlemail.com, johannes.rainer@eurac.edu
   affiliation: European Molecular Biology Laboratory, Meyerhofstrasse 1, 69117 Heidelberg/ EURAC Research, Viale Druso 1, 39100 BOlzano
package: MsQuality
      
output:  
    BiocStyle::html_document:
        toc_float: true
bibliography: MsQuality-citations.bib
vignette: >
    %\VignetteIndexEntry{QC for metabolomics and proteomics data}
    %\VignetteEngine{knitr::rmarkdown}
    %\VignetteKeywords{Mass Spectrometry, MS, Metabolomics, Proteomics, Visualization, QC}
    %\VignettePackage{MsQuality-vignette}
    %\VignetteEncoding{UTF-8}
---

```{r style, echo = FALSE, results = 'asis'}
BiocStyle::markdown()
```

```{r env, include=FALSE, echo=FALSE, cache=FALSE}
library("knitr")
opts_chunk$set(stop_on_error = 1L)
suppressPackageStartupMessages(library("MsQuality"))
suppressPackageStartupMessages(library("Spectra"))
```

# Introduction {#sec-intro}

Data quality assessment is an integral part of preparatory data analysis 
to ensure sound biological information retrieval. 

We present here the `MsQuality` package, which provides functionality to 
calculate quality metrics for mass spectrometry-derived, spectral data at the 
per-sample level. `MsQuality` relies on the 
[`mzQC`](https://github.com/HUPO-PSI/mzQC) framework of quality metrics defined 
by the Human Proteome Organization-Proteomics Standards Intitiative (HUPO-PSI). 
These metrics quantify the quality of spectral raw files using a controlled 
vocabulary. The package is especially addressed towards users that acquire 
mass spectrometry data on a large scale (e.g. data sets from clinical settings 
consisting of several thousands of samples): while it is easier to control 
for high-quality data acquisition in small-scale experiments, typically run
in one or few batches, clinical data sets are often acquired over longer 
time frames and are prone to higher technical variation that is often
unnoticed. `MsQuality` tries to address this problem by calculating metrics that
can be stored along the spectral data sets (raw files or feature-extracted 
data sets). `MsQuality`, thus, facilitates the tracking of shifts in data 
quality and quantifies the quality using multiple metrics. It should be thus 
easier to identify samples that are of low quality (high-number of missing 
values, termination of chromatographic runs, low instrument sensitivity, etc.).

We would like to note here that these metrics only give an indication of 
data quality, and, before removing indicated low-quality samples from the
analysis more advanced analytics, e.g. using the implemented functionality
and visualizations in the `MatrixQCvis` package, should be scrutinized. 
Also, data quality 
should always be regarded in the context of the sample type and experimental 
settings, i.e. quality metrics should always be compared with regard to the
sample type, experimental setup, instrumentation, etc..

The `MsQuality` package allows to calculate low-level quality metrics that 
require minimum information on mass spectrometry data: retention time, 
m/z values, and associated intensities.
The list included in the `mzQC` framework is excessive, also including 
metrics that rely on more high-level information, that might not be readily
accessible from .raw or .mzML files, e.g. pump pressure mean, or rely
on alignment results, e.g. retention time mean shift, signal-to-noise ratio,
precursor errors (ppm). 

The `MsQuality` package is built upon the `Spectra` and the `MsExperiment` 
package.
Metrics will be calculated based on the information stored in a 
`Spectra` object and the respective `dataOrigin` entries are used to 
distinguish between the mass spectral data of multiple samples.
The `MsExperiment` serves as a container to 
store the mass spectral data of multiple samples. `MsQuality` enables the user
to calculate quality metrics both on `Spectra` and `MsExperiment` objects.

In this vignette, we will (i) create some exemplary `Spectra` and `MsExperiment`
objects, (ii) calculate the quality metrics on these
data sets, and (iii) visualize some of the metrics.

## Alternative software for data quality assessment

Other `R` packages are available in Bioconductor that are able to assess 
the quality of mass spectrometry data: 

[`artMS`](https://bioconductor.org/packages/release/bioc/html/artMS.html)
uses MaxQuant output and enables to calculate several QC metrics, e.g.
correlation matrix for technical replicates, calculation of total sum of 
intensities in biological replicates, total peptide counts in biological 
replicates, charge state distribution of PSMs identified in each biological
replicates, or MS1 scan counts in each biological replicate.

[`MSstatsQC`](https://bioconductor.org/packages/release/bioc/html/MSstatsQC.html)
and the visualization tool
[`MSstatsQCgui`](https://bioconductor.org/packages/release/bioc/html/MSstatsQCgui.html)
require csv files in long format from spectral processing tools such as 
Skyline and Panorama autoQC or `MSnbase` objects. `MSstatsQC` enables to 
generate individual, moving range, cumulative sum for mean, and/or
cumulative sum for variability control charts for each metric. Metrics
can be any kind of user-defined metric stored in the data columns for a given
peptide, e.g. retention time and peak area.

[`MQmetrics`](https://bioconductor.org/packages/release/bioc/html/MQmetrics.html)
provides a pipeline to analyze the quality of proteomics data sets from
MaxQuant files and focuses on proteomics-/MaxQuant-specific metrics, e.g.
proteins identified, peptides identified, or proteins versus 
peptide/protein ratio.

[`MatrixQCvis`](https://bioconductor.org/packages/release/bioc/html/MatrixQCvis.html)
provides an interactive shiny-based interface to assess data quality at various
processing steps (normalization, transformation, batch correction, and 
imputation) of rectangular matrices. The package includes several diagnostic
plots and metrics such as barplots of intensity distributions, plots to 
visualize drifts, MA plots and Hoeffding's D value calculation, and 
dimension reduction plots and provides specific tools to analyze data sets 
containing missing values as commonly observed in mass spectrometry.

[`proBatch`](https://bioconductor.org/packages/release/bioc/html/proBatch.html)
enables to assess batch effects in (prote)omics data sets and corrects these
batch effects in subsequent steps. Several tools to visualize data quality are 
included in the `proBatch` packages, such as barplots of intensity 
distributions, cluster and heatmap analysis tools, and PCA dimension 
reduction plots. Additionally, `proBatch` enables to assess diagnostics at 
the feature level, e.g. peptides or spike-ins.

# Installation

To install this package, start `R` and enter:

```{r, eval = FALSE}
if (!requireNamespace("BiocManager", quietly = TRUE))
    install.packages("BiocManager")

if (!requireNamespace("remotes", quietly = TRUE))
    install.packages("remotes")

## to install from Bioconductor
BiocManager::install("MsQuality")

## to install from GitHub
BiocManager::install("tnaake/MsQuality")
```

This will install this package and all eventually missing dependencies.

# Questions and bugs {-}

`MsQuality` is currently under active development. If you discover any bugs, 
typos or develop ideas of improving `MsQuality` feel free to raise an issue 
via [GitHub](https://github.com/tnaake/MsQuality) or send a mail to the 
developer.

# Create `Spectra` and `MsExperiment` objects 

Load the `Spectra` package.

```{r load_Spectra}
library("Spectra")
library("MsExperiment")
library("MsQuality")
```

## Create `Spectra` and `MsExperiment` objects from mzML files

There are several options available to create a `Spectra` object. One way, as
outlined in the vignette of the `r BiocStyle::Biocpkg("Spectra")` package is 
by specifying the location of mass spectrometry raw files in `mzML`, `mzXML` or 
`CDF` format and using the `MsBackendMzR` backend. Here we load the example 
files from the `sciex` data set of the `msdata` package and create a `Spectra` 
object from the two provided `mzML` files. The example is taken from the 
`Spectra` vignette.

```{r spectra_sciex, eval = FALSE, echo = TRUE}
## this example is taken from the Spectra vignette
fls <- dir(system.file("sciex", package = "msdata"), full.names = TRUE)
sps_sciex <- Spectra(fls, backend = MsBackendMzR())
```

The data set consists of a single sample measured in two different 
injections to the same LC-MS setup. An empty instance of an
`MsExperiment` object is created and populated with information on the samples
by assigning data on the samples (`sampleData`), information on the 
`mzML` files (`MsExperimentFiles`) and spectral information (`spectra`). 
In a last step, using `linkSampleData`, the relationships between the samples
and the spectral information are defined.

```{r msexp_sciex, eval = FALSE, echo = TRUE}
## this example is taken from the Spectra vignette
lmse <- MsExperiment()
sd <- DataFrame(sample_id = c("QC1", "QC2"),
    sample_name = c("QC Pool", "QC Pool"),
    injection_idx = c(1, 3))
sampleData(lmse) <- sd
## add mzML files to the experiment
experimentFiles(lmse) <- MsExperimentFiles(mzML_files = fls)
## add the Spectra object to the experiment
spectra(lmse) <- sps_sciex
## use linkSampleData to establish and define relationships between sample 
## annotations and MS data
lmse <- linkSampleData(lmse, with = "experimentFiles.mzML_file",
    sampleIndex = c(1, 2), withIndex = c(1, 2))
```

## Create `Spectra` and `MsExperiment` objects from (feature-extracted) intensity tables

Another common approach is the creation of `Spectra` objects from a 
`DataFrame`s using the `MsBackendDataFrame` backend. 

We will use here the data set of @Lee2019, that contains metabolite level 
information measured by reverse phase liquid chromatography (RPLC) coupled 
to mass spectrometry and hydrophilic interaction liquid chromatography (HILIC) 
coupled to mass spectrometry 
(derived from the file `STables - rev1.xlsx` in the Supplementary Information).

In a separate step (see documentation for `Lee2019_meta_vals` and
`Lee2019`), we have created a list containing `Spectra` objects for
each samples (objects `sps_l_rplc` and `sps_l_hilic`) and `MsExperiment`
objects containing the data of all samples (objects `msexp_rplc` and
`msexp_hilic`). We will load here these objects:

```{r load_Lee2019, eval = TRUE, echo = TRUE}
data("Lee2019", package = "MsQuality")
```

The final data set contains `r length(unique(sps_rplc$dataOrigin))`
paired samples (i.e. `r length(unique(sps_rplc$dataOrigin))` samples derived 
from RPLC and `r length(unique(sps_hilic$dataOrigin))` samples derived 
from HILIC).

We will combine the `sps_rplc` and `sps_hilic` objects in the following and
calculate on this combined document the metrics. 

```{r combine_sps_rplc_sps_hilic, eval = TRUE, echo = TRUE}
sps_comb <- c(sps_rplc, sps_hilic)
```

The most important function to assess the data quality and to calculate the 
metrics is the `calculateMetrics` function. The function takes
a `Spectra` or `MsExperiment` object as input, a character vector of metrics
to be calculated, and, optionally a list of parameters passed to the 
quality metrics functions. 

# Calculating the quality metrics on `Spectra` and `MsExperiment` objects

Currently, the following metrics are included:
```{r}
qualityMetrics(msexp_rplc)
```

The following list gives a brief explanation for these metrics. Further 
information may be found at the 
[HUPO-PSI mzQC project page](https://github.com/HUPO-PSI/mzQC)  or in the 
respective help file for the quality metric (accessible by e.g. entering
`?rtDuration` to the R console). Currently, all quality metrics can be 
calculated for both `Spectra` and `MsExperiment` objects. 

- *rtDuration*, **RT duration** (QC:4000053), 
  "The retention time duration of the MS run in seconds, similar to the 
  highest scan time minus the lowest scan time." [PSI:QC];
- *rtOverTicQuantiles*, **RT over TIC quantile** (QC:4000054), 
  "The interval when the respective quantile of the TIC accumulates divided by 
  retention time duration. The number of quantiles observed is given by the 
  size of the tuple." [PSI:QC];
- *rtOverMsQuarters*, **MS1 quantiles RT fraction** (QC:4000055), 
  "The interval used for acquisition of the first, second, third, and fourth 
  quarter of all MS1 events divided by RT-Duration." [PSI:QC];
- *rtOverMsQuarters*, **MS1/MS2 quantiles RT fraction** (QC:4000055), 
  "The interval used for acquisition of the first, second, third, and fourth 
  quarter of all MS2 events divided by RT-Duration." [PSI:QC];
- *ticQuartileToQuartileLogRatio*, **MS1 TIC-change quartile ratios** 
  (MS:4000057), "The log ratios of successive TIC-change quartiles. The 
  TIC changes are the list of MS1 total ion current (TIC) value changes 
  from one to the next scan, produced when each MS1 TIC is subtracted 
  from the preceding MS1 TIC. The metric's value triplet represents the 
  log ratio of the TIC-change Q2 to Q1, Q3 to Q2, TIC-change-max to Q3" 
  [PSI:MS], `mode = "TIC_change"`;
- *ticQuartileToQuartileLogRatio*, **MS1 TIC quartile ratios** (MS:4000058),
  The log ratios of successive TIC quartiles. The metric's value triplet 
  represents the log ratios of TIC-Q2 to TIC-Q1, TIC-Q3 to TIC-Q2, TIC-max to 
  TIC-Q3." [PSI:MS], `mode = "TIC";
- *numberSpectra*, **Number of MS1 spectra** (QC:4000059), 
  "The number of MS1 events in the run." [PSI:QC];
- *numberSpectra*, **Number of MS2 spectra** (QC:4000060), 
  "The number of MS2 events in the run." [PSI:QC];
- *medianPrecursorMz*, **Precursor median m/z for IDs** (QC:4000065), 
  "Median m/z value for all identified peptides (unique ions) after 
  FDR." [PSI:QC];
- *rtIqr*, **Interquartile RT period for peptide identifications** (QC:4000072),
  "The interquartile retention time period, in seconds, for all peptide 
  identifications over the complete run." [PSI:QC];
- *rtIqrRate*, **Peptide identification rate of the interquartile RT period**
  (QC:4000073), 
  "The identification rate of peptides for the interquartile retention time 
  period, in peptides per second." [PSI:QC];
- *areaUnderTic*, **Area under TIC** (QC:4000077),
  "The area under the total ion chromatogram." [PSI:QC];
- *areaUnderTicRtQuantiles*, **Area under TIC RT quantiles** (QC:4000078),
  "The area under the total ion chromatogram of the retention time quantiles. 
  Number of quantiles are given by the n-tuple." [PSI:QC];
- *extentIdentifiedPrecursorIntensity*, 
  **Extent of identified precursor intensity** (QC:4000125),
  "Ratio of 95th over 5th percentile of precursor intensity for identified 
  peptides" [PSI:QC];
- *medianTicRtIqr*, **Median of TIC values in the RT range in which the middle 
  half of peptides are identified** (QC:4000130),
  "Median of TIC values in the RT range in which half of peptides are 
  identified (RT values of Q1 to Q3 of identifications)" [PSI:QC];
- *medianTicOfRtRange*, **Median of TIC values in the shortest RT range in 
  which half of the peptides are identified** (QC:4000132),
  "Median of TIC values in the shortest RT range in which half of the 
  peptides are identified"  [PSI:QC] 
- *mzAcquisitionRange*, **m/z acquisition range** (QC:4000138),
  "Upper and lower limit of m/z values at which spectra are recorded." [PSI:QC];
- *rtAcquisitionRange*, **Retention time acquisition range** (QC:4000139),
  "Upper and lower limit of time at which spectra are recorded." [PSI:QC];
- *precursorIntensityRange*, **Precursor intensity range** (QC:4000144),
  "Minimum and maximum precursor intensity recorded." [PSI:QC];
- *precursorIntensityQuartiles*, **Precursor intensity distribution 
  Q1, Q2, Q3** (QC:4000167), 
  "From the distribution of precursor intensities, the quartiles Q1, Q2, Q3" 
  [PSI:QC];
- *precursorIntensityQuartiles*, **Identified precursor intensity distribution 
  Q1, Q2, Q3** (QC:4000228), 
  "From the distribution of identified precursor intensities, the quartiles 
  Q1, Q2, Q3" [PSI:QC];
- *precursorIntensityQuartiles*, **Unidentified precursor intensity 
  distribution Q1, Q2, Q3** (QC:4000233), 
  "From the distribution of unidentified precursor intensities, the quartiles 
  Q1, Q2, Q3" [PSI:QC];
- *precursorIntensityMean*, **Precursor intensity distribution mean** 
  (QC:4000168), "From the distribution of precursor intensities, the mean." 
  [PSI:QC];
- *precursorIntensityMean*, **Identified precursor intensity distribution 
  mean** (QC:4000229), "From the distribution of identified precursor 
  intensities, the mean" [PSI:QC];
- *precursorIntensityMean*, **Unidentified precursor intensity distribution 
  mean** (QC:4000234), "From the distribution of unidentified precursor 
  intensities, the mean" [PSI:QC];
- *precursorIntensitySd*, **Precursor intensity distribution sigma**
  (QC:4000169), "From the distribution of precursor intensities, the sigma 
  value." [PSI:QC];
- *precursorIntensitySd*, **Identified precursor intensity distribution sigma**
  (QC:4000230), "From the distribution of identified precursor intensities, 
  the sigma value" [PSI:QC];
- *precursorIntensitySD*, **Unidentified precursor intensity distribution 
  sigma** (QC:400235), "From the distribution of unidentified precursor 
  intensities, the sigma value" [PSI:QC];
- *msSignal10xChange*, **MS1 signal jump (10x) count** (QC:4000172),
  "The count of MS1 signal jump (spectra sum) by a factor of ten or more (10x)
  between two subsequent scans" [PSI:QC];
- *msSignal10xChange*, **MS1 signal fall (10x) count** (QC:4000173),
  "The count of MS1 signal decline (spectra sum) by a factor of ten or more 
  (10x) between two subsequent scans" [PSI:QC];
- *RatioCharge1over2*, **Charged peptides ratio 1+ over 2+** (QC:4000174),
  "Ratio of 1+ peptide count over 2+ peptide count in identified spectra" 
  [PSI:QC];
- *RatioCharge1over2*, **Charged spectra ratio 1+ over 2+** (QC:4000179),
  "Ratio of 1+ spectra count over 2+ spectra count in all MS2" [PSI:QC];
- *RatioCharge3over2*, **Charged peptides ratio 3+ over 2+** (QC:4000175),
  "Ratio of 3+ peptide count over 2+ peptide count in identified spectra" 
  [PSI:QC];
- *RatioCharge3over2*, **Charged spectra ratio 3+ over 2+** (QC:4000180),
  "Ratio of 3+ peptide count over 2+ peptide count in all MS2" [PSI:QC];
- *RatioCharge4over2*, **Charged peptides ratio 4+ over 2+** (QC:4000176),
  "Ratio of 4+ peptide count  over 2+ peptide count in identified 
  spectra" [PSI:QC];
- *RatioCharge4over2*, **Charged spectra ratio 4+ over 2+** (QC:4000181),
  "Ratio of 4+ peptide count over 2+ peptide count in all MS2" [PSI:QC];
- *meanCharge*, **Mean charge in identified spectra** (QC:4000177),
  "Mean charge in identified spectra" [PSI:QC];
- *meanCharge*, **Mean precursor charge in all MS2** (QC:4000182),
  "Mean precursor charge in all MS2" [PSI:QC];
- *medianCharge*, **Median charge in identified spectra** (QC:4000178),
  "Median charge in identified spectra" [PSI:QC];
- *medianCharge*, **Median precursor charge in all MS2** (QC:4000183),
  "Median precursor charge in all MS2" [PSI:QC].

The most important function to assess the data quality and to calculate the 
metrics is the `calculateMetrics` function. The function takes
a `Spectra` or `MsExperiment` object as input, a character vector of metrics
to be calculated, and, optionally a list of parameters passed to the 
quality metrics functions. 

When passing a `Spectra`/`MsExperiment` object to the function, a `data.frame` 
returned by `calculateMetrics` with the metrics specified by the argument
`metrics`. By default, `qualityMetrics(object)` is taken to specify
the calculation of quality metrics. `calculateMetrics` also accepts a list
of parameters passed to the individual quality metrics functions. For 
each quality metrics functions, the relevant parameters are selected based on 
the accepted arguments. 

Additional arguments can be given to the quality metrics functions. 
For example, the function `ticQuartileToQuartileLogRatio` function has the 
arguments `relativeTo`, `mode`, and `msLevel`. `relativeTo` specifies to which 
quantile the log TIC quantile is relatively related to (either to the 1st 
quantile or the respective previous one). `mode` (either `"TIC_change"` or 
`"TIC"`) specifies if the quantiles are taken from the changes between TICs 
of scan events or the TICs directly. One `Spectra`/`MsExperiment` object may 
also contain more than one `msLevel`, e.g. if it also contains information on
MS$^2$ or MS$^3$ features. If the user adds the arguments 
`relativeTo = "Q1", mode = "TIC", msLevel = c(1L, 2L))`, 
`ticQuartileToQuartileLogRatio` is run with the parameter combinations 
`relativeTo = "Q1", mode = "TIC", msLevel = c(1L, 2L)`.

The  results based on these parameter combinations are returned and the 
used parameters are returned as attributes to the returned vector. 

Here, we would like to calculate the metrics of **all** included quality 
metrics functions (`qualityMetrics(object)`) and additionally pass the 
parameter `relativeTo = "Q1"` and `relativeTo = "previous"`. For computational
reasons, we will restrict the calculation of the metrics to the first  
sample and to RPLC samples.

```{r calculateMetrics}
## subset the Spectra objects
sps_comb_subset <- sps_comb[grep("Sample.1_", sps_comb$dataOrigin), ]

## for RPLC and HILIC
metrics_sps_Q1 <- calculateMetrics(object = sps_comb_subset,
    metrics = qualityMetrics(sps_comb_subset),
    relativeTo = "Q1", msLevel = 1L)
metrics_sps_Q1
metrics_sps_previous <- calculateMetrics(object = sps_comb_subset,
    metrics = qualityMetrics(sps_comb_subset),
    relativeTo = "previous", msLevel = 1L)
metrics_sps_previous
```

Alternatively, an `MsExperiment` object might be passed to 
`calculateMetrics`. The function will iterate over the samples (referring
to rows in `sampleData(msexp))`) and calculate the quality metrics on the
corresponding `Spectra`s.

There are in total `r length(msexp_hilic)` samples 
respectively in the objects `msexp_rplc` and `msexp_hilic`. To improve 
the visualization and interpretability, we will only calculate the metrics 
from the first 20 of these samples.

```{r metrics_rplc_hilic_lee2019}
## subset the MsExperiment objects
msexp_rplc_subset <- msexp_rplc[1:20]
msexp_hilic_subset <- msexp_hilic[1:20]

## define metrics
metrics_sps <- c("rtDuration", "rtOverTicQuantiles", "rtOverMsQuarters",
    "ticQuartileToQuartileLogRatio", "numberSpectra", "medianPrecursorMz",
    "rtIqr", "rtIqrRate", "areaUnderTic")

## for RPLC-derived MsExperiment
metrics_rplc_msexp <- calculateMetrics(object = msexp_rplc_subset,
    metrics = qualityMetrics(msexp_rplc_subset),
    relativeTo = "Q1", msLevel = 1L)

## for HILIC-derived MsExperiment
metrics_hilic_msexp <- calculateMetrics(object = msexp_hilic_subset,
    metrics = qualityMetrics(msexp_hilic_subset),
    relativeTo = "Q1", msLevel = 1L)
```

When passing an `MsExperiment` object to `calculateMetrics` a `data.frame`
object is returned with the samples (derived from the rownames of 
`sampleData(msexp)`) in the rows and the metrics in columns. 

We will show here the objects `metrics_rplc_msexp` and `metrics_hilic_msexp`
```{r paged_table_metrics, eval = TRUE, echo = FALSE}
print("metrics_rplc_msexp")
rmarkdown::paged_table(as.data.frame(metrics_rplc_msexp))
print("metrics_hilic_msexp")
rmarkdown::paged_table(as.data.frame(metrics_hilic_msexp))
```

# Visualizing the results

The quality metrics can be most easily compared when graphically visualized.

The `MsQuality` package offers the possibility to graphically display the 
metrics using the `plotMetric` and `shinyMsQuality` functions. The 
`plotMetric` function will create one plot based on a single metric. 
`shinyMsQuality`, on the other hand, opens a shiny application that allows
to browse through all the metrics stored in the object.

As a way of example, we will plot here the number of features. A high number
of missing features might indicate low data quality, however, also different
sample types might exhibit contrasting number of detected features.
As a general rule, only samples of the same type should be compared to
adjust for sample type-specific effects.

```{r}
metrics_msexp <- rbind(metrics_rplc_msexp, metrics_hilic_msexp)
plotMetric(qc = metrics_msexp, metric = "numberSpectra")
```

Similarly, we are able to display the area under the TIC for the retention
time quantiles. This plot gives information on the perceived signal (TIC) for
the differnt retention time quantiles and could indicate drifts or
interruptions of sensitivity during the run.

```{r}
plotMetric(qc = metrics_msexp, metric = "ticQuartileToQuartileLogRatio")
```

Alternatively, to browse through all metrics that were calculated in an 
interactive way, we can use the `shinyMsQuality` function. 

```{r eval = FALSE}
shinyMsQuality(qc = metrics_msexp)
```

# Appendix {-}

## Session information {-}

All software and respective versions to build this vignette are listed here:

```{r session,eval=TRUE, echo=FALSE}
sessionInfo()
```

## References