Chapter 7 Mass spectrometry

7.1 Setup

Copy link

Figure 7.1: The Spectra package.

We are going to use the Bioconductor (⊕Huber et al. 2015Huber, W, V J Carey, R Gentleman, S Anders, M Carlson, B S Carvalho, H C Bravo, et al. 2015. “Orchestrating High-Throughput Genomic Analysis with Bioconductor.” Nat Methods 12 (2): 115–21. https://doi.org/10.1038/nmeth.3252.) Spectra package, which can be install with the BiocManager package, available from CRAN. We will also make use of some test data from the msdata package. Identification data will be handled by the PSMatch package.

BiocManager::install("msdata")
BiocManager::install("ProtGenerics")
BiocManager::install("Spectra")
BiocManager::install("PSMatch")

7.2 How does mass spectrometry work?

Copy link

Mass spectrometry (MS) is a technology that separates charged molecules (ions) based on their mass to charge ratio (M/Z). It is often coupled to chromatography (liquid LC, but can also be gas-based GC). The time an analytes takes to elute from the chromatography column is the retention time.

Figure 7.2: A chromatogram, illustrating the total amount of analytes over the retention time.

An mass spectrometer is composed of three components:

1. The source, that ionises the molecules: examples are Matrix-assisted laser desorption/ionisation (MALDI) or electrospray ionisation. (ESI)
2. The analyser, that separates the ions: Time of flight (TOF) or Orbitrap.
3. The detector that quantifies the ions.

When using mass spectrometry for proteomics, the proteins are first digested with a protease such as trypsin. In mass shotgun proteomics, the analytes assayed in the mass spectrometer are peptides.

Often, ions are subjected to more than a single MS round. After a first round of separation, the peaks in the spectra, called MS1 spectra, represent peptides. At this stage, the only information we possess about these peptides are their retention time and their mass-to-charge (we can also infer their charge be inspecting their isotopic envelope, i.e the peaks of the individual isotopes, see below), which is not enough to infer their identify (i.e. their sequence).

In MSMS (or MS2), the settings of the mass spectrometer are set automatically to select a certain number of MS1 peaks (for example 20). Once a narrow M/Z range has been selected (corresponding to one high-intensity peak, a peptide, and some background noise), it is fragmented (using for example collision-induced dissociation (CID), higher energy collisional dissociation (HCD) or electron-transfer dissociation (ETD)). The fragment ions are then themselves separated in the analyser to produce a MS2 spectrum. The unique fragment ion pattern can then be used to infer the peptide sequence using de novo sequencing (when the spectrum is of high enough quality) of using a search engine such as, for example Mascot, MSGF+, …, that will match the observed, experimental spectrum to theoratical spectra (see details below).

Figure 7.3: Schematics of a mass spectrometer and two rounds of MS.

The animation below show how 25 ions different ions (i.e. having different M/Z values) are separated throughout the MS analysis and are eventually detected (i.e. quantified). The final frame shows the hypothetical spectrum.

Figure 7.4: Separation and detection of ions in a mass spectrometer.

See also the Johannes Rainer’s animations in the SpectraTutorials vignette.

The figures below illustrate the two rounds of MS. The spectrum on the left is an MS1 spectrum acquired after 21 minutes and 3 seconds of elution. 10 peaks, highlited by dotted vertical lines, were selected for MS2 analysis. The peak at M/Z 460.79 (488.8) is highlighted by a red (orange) vertical line on the MS1 spectrum and the fragment spectra are shown on the MS2 spectrum on the top (bottom) right figure.

Figure 7.5: Parent ions in the MS1 spectrum (left) and two sected fragment ions MS2 spectra (right).

The figures below represent the 3 dimensions of MS data: a set of spectra (M/Z and intensity) of retention time, as well as the interleaved nature of MS1 and MS2 (and there could be more levels) data.

Figure 7.6: MS1 spectra (blue) over retention time (left). MS2 spectra (pink) interleaved between two MS1 spectra (right),

7.3 Reading and accessing MS data

Copy link

Let’s start by accessing an MS raw data file from the msdata package. The proteomics() function checks the package content for a file matching the "2014" pattern and returns its full name.

basename(f <- msdata::proteomics(pattern = "2014", full.names = TRUE))

## [1] "TMT_Erwinia_1uLSike_Top10HCD_isol2_45stepped_60min_01-20141210.mzML.gz"

There exist a set of community-maintained file formats for mass spectrometry data. The most widely used formats are mzML and mzXML, both relying on the XML markup language. These two formats, as well as others, can be read into R as Spectra objects:

library("Spectra")
ms <- Spectra(f)
ms

## MSn data (Spectra) with 7534 spectra in a MsBackendMzR backend:
##        msLevel     rtime scanIndex
##      <integer> <numeric> <integer>
## 1            1    0.4584         1
## 2            1    0.9725         2
## 3            1    1.8524         3
## 4            1    2.7424         4
## 5            1    3.6124         5
## ...        ...       ...       ...
## 7530         2   3600.47      7530
## 7531         2   3600.83      7531
## 7532         2   3601.18      7532
## 7533         2   3601.57      7533
## 7534         2   3601.98      7534
##  ... 33 more variables/columns.
## 
## file(s):
## TMT_Erwinia_1uLSike_Top10HCD_isol2_45stepped_60min_01-20141210.mzML.gz

Once a Spectra object has been created, it can be queried and manipulated using a whole set of accessor functions, documented in the Spectra-class documentation.

Spectra objects can be subset with the [ operator. In addition to the raw data (the M/Z and intensity values) that compose the individual scans, the annotation of each spectrum is available as a single and large table called spectraData, composed of a set of spectraVariables columns.

Figure 7.7: Schematic representation of what is referred to by MS data: a collection of spectra and a table containing spectrum-level annotations.

The figure below show is an illustration of how mass spectrometry works:

1. The chromatogram at the top display to total ion current along the retention time. The vertical line identifies one scan in particular at retention time 1800.68 seconds (the 2807th scan).

2. The spectra on the second line represent the full MS1 spectrum marked by the red line. The vertical lines identify the 10 precursor ions that where selected for MS2 analysis. The zoomed in on the right shows one specific precursor peak.

3. The MS2 spectra displayed along the two rows at the bottom are those resulting from the fragmentation of the 10 precursor peaks identified by the vertical bars above.

with(spectraData(filterMsLevel(ms, 1)),
     plot(rtime, totIonCurrent, type = "l"))
abline(v = rtime(ms)[2807], col = "red")

ms_2 <- filterPrecursorScan(ms, 2807)
plotSpectra(ms[2807], xlim = c(400, 1000))
abline(v = precursorMz(ms_2)[-1], col = "grey")
abline(v = precursorMz(ms_2)[2], col = "red")

plotSpectra(ms[2807], xlim = c(521.2, 522.5), type = "l")
abline(v = precursorMz(ms_2)[2], col = "red")

plotSpectra(ms_2[-1])

7.4 Identification: matching spectra to peptides

Copy link

The raw data is still a long way of obtaining biologically relevant proteomics data. The first step is to identify the peptides that have been acquired in the MS. Peptide identification works by comparing expected and observed spectra. As shown below, when a precursor peptide ion is fragmented in a CID cell, it breaks at specific bonds, producing sets of peaks (a, b, c and x, y, z) that can be predicted.

Figure 7.8: Peptide fragmentation.

It is thus possible to calculate the expected set of fagment peaks for a given peptide, such as THSQEEMQHMQR below and match these theoretical fragments to those observed in MS2 spectra.

library("PSMatch")
calculateFragments("THSQEEMQHMQR")

##           mz ion type pos z         seq
## 1   102.0550  b1    b   1 1           T
## 2   239.1139  b2    b   2 1          TH
## 3   326.1459  b3    b   3 1         THS
## 4   454.2045  b4    b   4 1        THSQ
## 5   583.2471  b5    b   5 1       THSQE
## 6   712.2897  b6    b   6 1      THSQEE
## 7   843.3301  b7    b   7 1     THSQEEM
## 8   971.3887  b8    b   8 1    THSQEEMQ
## 9  1108.4476  b9    b   9 1   THSQEEMQH
## 10 1239.4881 b10    b  10 1  THSQEEMQHM
## 11 1367.5467 b11    b  11 1 THSQEEMQHMQ
## 12  175.1190  y1    y   1 1           R
## 13  303.1775  y2    y   2 1          QR
## 14  434.2180  y3    y   3 1         MQR
## 15  571.2769  y4    y   4 1        HMQR
## 16  699.3355  y5    y   5 1       QHMQR
##  [ reached 'max' / getOption("max.print") -- omitted 42 rows ]

In a full experiment, all possible peptides from the proteome of interest (such as databases that can be downloaded from the UniProt site¹³) are compared to the millions of observed spectra.

The matching between millions of observed and possible spectra causes real challenges due to the large search space and the risk of false positives. See (⊕Käll et al. 2008Käll, L, J D Storey, M J MacCoss, and W S Noble. 2008. “Posterior Error Probabilities and False Discovery Rates: Two Sides of the Same Coin.” J Proteome Res 7 (1): 40–44. https://doi.org/10.1021/pr700739d.) for further reading.

7.4.1 Handling identification results

Copy link

Identification results are also stored in an XML-based file format, called mzIdentML (or mzId for short). Below, we access such a file available in the msdata package and load its content into R with the PSMs::PSM() function.

basename(idf <- msdata::ident(full.names = TRUE))

## [1] "TMT_Erwinia_1uLSike_Top10HCD_isol2_45stepped_60min_01-20141210.mzid"

id <- PSM(idf)
head(id)

## PSM with 6 rows and 35 columns.
## names(35): sequence spectrumID ... subReplacementResidue subLocation

► Question

How many of PSM entries originate from the real and decoy databases?

► Solution

table(id$isDecoy)

## 
## FALSE  TRUE 
##  2906  2896

► Question

Visualise the MS.GF.RawScore densitues for the real and decoy hits. Interprete the figures.

► Solution

library("ggplot2")
data.frame(id) %>%
    ggplot(aes(x = MS.GF.RawScore, colour = isDecoy)) +
    geom_density()

Before proceeding, we want to only focus on relevant peptides. The filterPSMs() function will remove decoy peptides, only keep PSMs of rank 1 (i.e. the best respective matches), and remove peptides that match multiple proteins¹⁴.

id <- filterPSMs(id)

## Starting with 5802 PSMs:

## Removed 2896 decoy hits.

## Removed 155 PSMs with rank > 1.

## Removed 85 shared peptides.

## 2666 PSMs left.

For simplicity, we are also going to ignore the location of the peptide modifications by removing the modLocation column.

id$modLocation <- NULL
id <- unique(id)

Now that we have matching raw data (the ms variable) and identification data (the id variable), it is possible to join the latter to the former’s spectra data:

ms <- joinSpectraData(ms, id,
                      by.x = "spectrumId",
                      by.y = "spectrumID")
ms

## MSn data (Spectra) with 7534 spectra in a MsBackendMzR backend:
##        msLevel     rtime scanIndex
##      <integer> <numeric> <integer>
## 1            1    0.4584         1
## 2            1    0.9725         2
## 3            1    1.8524         3
## 4            1    2.7424         4
## 5            1    3.6124         5
## ...        ...       ...       ...
## 7530         2   3600.47      7530
## 7531         2   3600.83      7531
## 7532         2   3601.18      7532
## 7533         2   3601.57      7533
## 7534         2   3601.98      7534
##  ... 66 more variables/columns.
## 
## file(s):
## TMT_Erwinia_1uLSike_Top10HCD_isol2_45stepped_60min_01-20141210.mzML.gz

We are going to conlude this part by annotating an MS2 spectrum with the calculated peptide fragment.

► Question

• Choose two MS2 spectra that have an MS.GF.RawScore greater than 100 and are matched to the same peptide sequence.

• Then visualise them annotating their fragment ion with its fragment ions by passing the addFragments function to plotSpectra’s labels argument.

► Solution

seq <- as_tibble(spectraData(ms)) %>%
    filter(`MS.GF.RawScore` > 100) %>%
    group_by(sequence) %>%
    summarise(n = n()) %>%
    filter(n == 2) %>%
    head(1) %>%
    pull(sequence)

i <- which(ms$sequence == seq)

plotSpectra(ms[i], labels = addFragments,
            labelPos = 3, labelCol = "orange",
            main = ms[i]$sequence)

Figure 7.9: Two MS2 spectra that were matched to the TKGLNVMQNLLTAHPDVQAVFAQNDEMALGALR peptide.

For a more direct comparision, the two spectra can also be represented on the same graph pointing upwards and downwards:

plotSpectraMirror(ms[i[1]], ms[i[2]],
                  labels = addFragments,
                  labelPos = 3, labelCol = "orange")

Figure 7.10: Mirror representation of to matches to the TKGLNVMQNLLTAHPDVQAVFAQNDEMALGALR peptide.

From the list of identified peptides, it is then necessary to infer the most provable proteins that were present in the biological sample. Protein inference is a difficult task, as peptides often match multiple proteins (either different isoforms or proteins stemming from different gene but with identical domains), which leads to the definition of protein groups, i.e. sets of proteins that can’t be distinguished with the set of identified peptides at hand. See (⊕Nesvizhskii and Aebersold 2005Nesvizhskii, A I, and R Aebersold. 2005. “Interpretation of Shotgun Proteomic Data: The Protein Inference Problem.” Mol Cell Proteomics 4 (10): 1419–40. https://doi.org/10.1074/mcp.R500012-MCP200.) for further reading.

7.5 Quantitative proteomics

Copy link

The last step of MS data processing is to quantify peptide abundances in the biological samples. The table below summarises the different possibilites depending whether the proteins or peptides are labelled, and whether the quantitation is performed in MS1 or MS2.

7.5.1 Label-free MS2: Spectral counting

Copy link

In spectral counting, on simply counts the number of quantified peptides that are assigned to a protein.

Figure 7.11: Spectral counting. Figure from the Pbase package.

7.5.2 Labelled MS2: Isobaric tagging

Copy link

Isobaric tagging refers to the labelling using isobaric tags, i.e. chemical tags that have the same mass and hence can’t be distinguish by the spectrometer. The peptides of different samples (4, 6, 10, 11 or 16) are labelled with different tags and combined prior to mass spectrometry acquisition. Given that they are isobaric, all identical peptides, irrespective of the tag and this the sample of origin, are co-analysed, up to fragmentation prior to MS2 analysis. During fragmentation, the isobaric tags fall of, fragment themselves, and result in a set of sample specific peaks. These specific peaks can be used to infer sample-specific quantitation, while the rest of the MS2 spectrum is used for identification.

Figure 7.12: iTRAQ 4-plex isobaric tagging. Tandem Mass Tags (TMT) offer up to 16 tags.

7.5.3 Label-free MS1: extracted ion chromatograms

Copy link

In label-free quantitation, the precursor peaks that match an identified peptide are integrated of retention time and the area under that extracted ion chromatogram is used to quantify that peptide in that sample.

Figure 7.13: Label-free quantitation. Figure credit Johannes Rainer.

7.5.4 Labelled MS1: SILAC

Copy link

In SILAc quantitation, sample are grown in a medium that contains heavy amino acids (typically arginine and lysine). All proteins gown in this heavy growth medium contain the heavy form of these amino acids. Two samples, one grown in heavy medium, and one grown in normal (light) medium are then combined and analysed together. The heavy peptides precursor peaks are systematically shifted compared to the light ones, and the ratio between the height of a heavy and light peaks can be used to calculate peptide and protein fold-changes.

Figure 7.14: Silac quantitation. Figure credit Wikimedia Commons.

These different quantitation techniques come with their respective benefits and distinct challenges, such as large quantities of raw data processing, data transformation and normalisation, missing values, and different underlying statistical models for the quantitative data (count data for spectral counting, continuous data for the others).

The next chapter will focus on the handling and processing of quantitative proteomics data.