This vignette explains the main concepts for data-processing and statistical analysis of (DIA) proteomics data using msqrob2. To illustrate these concepts, we will use using a publicly available spike-in study published by Staes et al. (Staes et al. 2024). They spiked digested UPS proteins in a yeast digested background at the following ratio’s (yeast:ups ratio 10:1, 10:2, 10:4, 10:8, 10:10). Here we will use a subset of the data, i.e. dilutions 10:2, 10:4 and 10:8. We will use output of the search engine Spectronaut. The main search output for this Spectronaut version was stored in a report.parquet file.

Spectronaut provides multiple quantifications, e.g. derived from the MS1 or MS2 spectra, and at precursor or protein (protein group) level. The term ‘precursor’ refers to a charged peptide species and is the basic unit of identification and quantification in DIA. Hence, in the context of DIA we refer to a precursor table, instead of to a PSM table in DDA.

Examples of different quantities are:

  • raw MS1 quantity: FG_MS1RawQuantity, normalised MS1 quantity: FG_MS1Quantity, MS2 Precursor quantities: FG_MS2RawQuantity, Normalised MS2 Precursor quantities: FG_MS2Quantity, etc., which are all at the precursor level.
  • MS2 based protein(protein group)-level summaries are stored in PG_MaxLFQ.

In this vignette, we will use the FG_MS2RawQuantity column so as to illustrate all steps. If you prefer to start from the FG_MS2Quantity intensities, you can skip the normalisation step as the normalisation already has been done by Spectronaut internally.

1 Load packages

We load the msqrob2 package, along with additional packages for data manipulation and visualisation.

suppressPackageStartupMessages({
library("QFeatures")  
library("dplyr") 
library("tidyr")
library("ggplot2")
library("msqrob2")    
library("stringr")
library("ExploreModelMatrix")
library("kableExtra")
library("ComplexHeatmap")
library("scater")
})

2 Data

2.1 Precursor table

We load the output from Spectronaut parquet file.

precursorFile <- "https://github.com/statOmics/msqrob2data/raw/refs/heads/main/dia/spikeinStaes/spikein248-staesetal2024-spectronaut.parquet"

We can import the report.parquet file using the read_parquet function from the arrow package. Note, that older versions of Spectronaut store the output as report.tsv, which we import using the fread function. Both read_parquet and fread can import data directly from the web as well as data from local files on your computer.

precursors <- arrow::read_parquet(precursorFile) # function from the arrow package
## For older versions of Spectronaut, where the results are stored as tsv files. 
## Note that the precursorFile then would point to "report.tsv" 
## precursors <- data.table::fread(precursorFile) 
knitr::kable(head(precursors))
R_FileName PG_Genes PG_ProteinAccessions PG_ProteinGroups PG_ProteinNames PG_Organisms PG_NrOfPrecursorsIdentified PG_NrOfPrecursorsIdentified_(Experiment-wide) PG_Qvalue PG_QValue_(Run-Wise) PG_MS1Quantity PG_MS2Quantity PG_Quantity PEP_StrippedSequence EG_ModifiedSequence EG_PrecursorId FG_Charge EG_IsDecoy PEP_IsProteotypic PEP_MS1Quantity PEP_MS2Quantity PEP_Quantity PEP_UsedForProteinGroupQuantity EG_IonMobility EG_ApexRT FG_CalibratedMz EG_Qvalue EG_IsImputed FG_Qvalue FG_PrecursorSignalToNoise FG_SignalToNoise FG_ShapeQualityScore FG_ShapeQualityScore_(MS1) FG_ShapeQualityScore_(MS2) FG_MS1Quantity FG_MS1RawQuantity FG_MS2Quantity FG_MS2RawQuantity EG_NormalizationFactor
B000254_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio02_DIA TY4B-P;TY4B-H;TY4A-J;TY4B-J;TY4A-H A0A0B7P3V8;P0C2J7;P47023;P47024;Q6Q5P6 A0A0B7P3V8;P0C2J7;P47023;P47024;Q6Q5P6 YP41B_YEAST;YH41B_YEAST;YJ41A_YEAST;YJ41B_YEAST;YH41A_YEAST Saccharomyces cerevisiae (strain ATCC 204508 / S288c) 1 1 0 0.0000777 1737294.12 23325.660 23325.660 IVDLVEK IVDLVEK IVDLVEK.2 2 FALSE False 1737294.125 23325.660 23325.660 TRUE 0.7611104 76.33237 408.2469 1,123548054819927E-05 FALSE 0.0000112 NaN 13.076940 0.5988563 0.8960734 0.4609673 1737294.125 1330965.38 23325.660 17870.115 1.3052888
B000254_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio02_DIA YPR010C-A A5Z2X5 A5Z2X5 YP010_YEAST Saccharomyces cerevisiae (strain ATCC 204508 / S288c) 1 1 0 0.0000000 279530.47 15254.879 15254.879 LTGNPELSSLDEVLAK LTGNPELSSLDEVLAK LTGNPELSSLDEVLAK.2 2 FALSE True 279530.469 15254.879 15254.879 TRUE 1.0992813 117.79333 843.4530 1,8731018963979458E-07 FALSE 0.0000002 NaN 18.562664 0.6349962 0.4518091 0.8761904 279530.469 274669.47 15254.879 14989.599 1.0176976
B000254_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio02_DIA STE2 D6VTK4 D6VTK4 STE2_YEAST Saccharomyces cerevisiae (strain ATCC 204508 / S288c) 3 3 0 0.0001036 36324.98 1596.699 1596.699 TNTITSDFTTSTDR TNTITSDFTTSTDR TNTITSDFTTSTDR.2 2 FALSE True 30351.500 1114.846 1114.846 TRUE 1.0337754 80.53452 780.3642 1,680149029187513E-05 FALSE 0.0000168 NaN 5.160248 0.3816292 0.5499600 0.3728411 30351.500 29111.75 1114.846 1069.309 1.0425860
B000254_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio02_DIA STE2 D6VTK4 D6VTK4 STE2_YEAST Saccharomyces cerevisiae (strain ATCC 204508 / S288c) 3 3 0 0.0001036 36324.98 1596.699 1596.699 LYDLYPR LYDLYPR LYDLYPR.2 2 FALSE True 7748.781 1316.819 1316.819 TRUE 0.8341860 88.16772 470.2505 0,0016666728759980882 FALSE 0.0016667 NaN 2.409254 0.3015331 0.3229307 0.3912955 7748.781 8079.68 1316.819 1373.051 0.9590456
B000254_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio02_DIA STE2 D6VTK4 D6VTK4 STE2_YEAST Saccharomyces cerevisiae (strain ATCC 204508 / S288c) 3 3 0 0.0001036 36324.98 1596.699 1596.699 TFVSETADDIEK TFVSETADDIEK TFVSETADDIEK.2 2 FALSE True 71964.930 1523.927 1523.927 TRUE 0.9687490 85.28220 677.8233 2,4746665476584473E-05 FALSE 0.0000247 NaN 1.662843 0.4326174 0.6393431 0.3356901 71964.930 75014.59 1523.927 1588.507 0.9593459
B000254_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio02_DIA CET1 O13297 O13297 CET1_YEAST Saccharomyces cerevisiae (strain ATCC 204508 / S288c) 14 15 0 0.0000000 221288.70 12164.803 12164.803 IDPSFLNIIPDDDLTK IDPSFLNIIPDDDLTK IDPSFLNIIPDDDLTK.2 2 FALSE True 140974.625 17137.367 17137.367 TRUE 1.1310339 129.05934 908.4742 1,4254762712598077E-05 FALSE 0.0000143 NaN 10.028002 0.6728779 0.6650686 0.6714950 140974.625 149003.81 17137.367 18113.422 0.9461142

Each row in the precursor data table is in “long format” and contains information about one precursor in a specific run (the table below shows the first 6 rows). The columns contains various descriptors about the precursor, such as its sequence, its charge, run, etc.

We select variables to reduce the memory footprint. (was already done to store the output in a small parquet file). We also provided the equivalent names of the variables in DIA-NN in comments.

precursors <- precursors |> 
  select(
         R_FileName, #Run, 
         EG_PrecursorId, #PrecursoR_Id, 
         EG_ModifiedSequence, #Modified.Sequence, 
         PEP_StrippedSequence, #Stripped.Sequence, 
         FG_Charge, #PrecursoR_Charge, 
         PG_ProteinGroups, #Protein.Group, 
         PG_ProteinNames, #Protein.Names,
         PG_ProteinAccessions, #Protein.Ids,
         PG_Genes, #Genes,
         FG_MS2RawQuantity, #PrecursoR_Quantity,
         FG_MS2Quantity, #PrecursoR_Normalised,
         FG_MS1RawQuantity, #Ms1.Area,
         FG_MS1Quantity, #Ms1.Normalised #No real DIA-NN counterpart 
         EG_NormalizationFactor, #Normalisation.Factor
         EG_Qvalue, #Q.Value, 
         #No spectronaut counterpart #Lib.Q.Value,
         PG_Qvalue, # PG_Q.Value,
         #No spectronaut counterpart #Lib.PG_Q.Value
         PEP_IsProteotypic, #Proteotypic,
         EG_IsDecoy, #Decoy,
         EG_ApexRT,#RT
         EG_IsImputed)

Unlike for real experiments, we known the ground truth for this dataset: UPS proteins were spiked in at different concentrations and are thus differentially abundant (DA, spiked in). Yeast proteins consist of the background proteome that is the same in all samples and are thus non-DA.

precursors <- precursors |> 
  mutate(species = grepl(pattern = "ups",PG_ProteinGroups) |> 
           as.factor() |>
           recode("TRUE"="ups","FALSE" = "yeast"))
precursors |> 
  pull(species) |>
  table()
## 
##  yeast    ups 
## 391266   7281

There is a problem with the encoding of some variables:

  • EG_Qvalue is of type character while it should be a numeric.
  • PEP_IsProteotypic is of type character and not a logical.
precursors$EG_Qvalue |> head()
## [1] "1,123548054819927E-05"  "1,8731018963979458E-07" "1,680149029187513E-05" 
## [4] "0,0016666728759980882"  "2,4746665476584473E-05" "1,4254762712598077E-05"
precursors$PEP_IsProteotypic |> head()
## [1] "False" "True"  "True"  "True"  "True"  "True"

We replace , in . in EG_Qvalue and cast it into a double and we cast PEP_IsProteotypic to a logical.

precursors <- precursors |> 
  mutate(EG_Qvalue = sub(pattern = ",", 
                         replacement = ".", 
                         EG_Qvalue) |> as.double(),
         PEP_IsProteotypic = as.logical(PEP_IsProteotypic))

precursors$EG_Qvalue |> head()
## [1] 1.123548e-05 1.873102e-07 1.680149e-05 1.666673e-03 2.474667e-05
## [6] 1.425476e-05
precursors$PEP_IsProteotypic |> head()
## [1] FALSE  TRUE  TRUE  TRUE  TRUE  TRUE

Quick check on distribution of intensities.

precursors |> 
  ggplot(aes(x = log2(FG_MS2RawQuantity))) +
  geom_density()

precursors |> 
  ggplot(aes(x = log2(FG_MS2RawQuantity))) +
  geom_density() + 
  facet_wrap(~EG_IsImputed)

Note, that imputation has occurred. We prefer to avoid imputation and will filter the imputed intensities out during the preprocessing.

2.2 Sample annotation table

The sample annotation table) is not available and can be generated from the run labels, as the researchers included information on the design in the filenames.

We will make a new data frame with the annotation.

  1. We first generate variable runCol, that gives an overview of the different runs in the experiment, i.e. the unique names in the column Run of the report file. This is a mandatory column in the annotation file that is required by the readQFeatures function that will be used to generate a QFeatures object with the quant data.

  2. Next we generate variable sampleId, to pinpoint the different samples in the dataset. This can be extracted from the run names as the researchers have stored information on the meta data in the file names for each run. E.g. for run B000282_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio04_DIA_2 this is the ratio, i.e. ratio04, and the replicate, i.e. _2 after DIA. Note, that the first replicate has no number.

    1. We first replace the redundant pattern “DIA” with an empty character string.
    2. Next we split the strings according to the pattern “UPS2” using the strsplit function and
    3. keep the right part of the string. We do this by looping over the list from strplit with an sapply loop, which takes the output of b. as its first argument, uses the function ‘[’ to subset vector of strings in each list element and uses an optional argument ‘2’ to select the second string of the vector (right part).

  3. We add the variable ratio by parsing the variable stringId and

    1. replacing pattern “ratio” by an empty string using the gsub function
    2. splitting the output of gsub according to pattern "_" and
    3. keeping the left part of the string (first element of the vector of strings in each list item of the substr output)
    4. converting the output of c to an integer
    5. converting the ratio into a factor

  4. We add the variable rep by parsing the variable stringId, and

    1. splitting it according to pattern "_",
    2. keeping the right part of the string (second element of the vector of strings in each list item of the substr output)
    3. replacing NA by 1, as for the first replicate no number was added to the filename
    4. converting it into a factor
annot <- data.frame(runCol = precursors |> 
                       pull(R_FileName) |> 
                       unique() # 1. 
                   ) |>
    mutate(sampleId = gsub(x = runCol, pattern = "_DIA", replacement = "") |> #2.a
               str_split("UPS2_") |> #2.b
               sapply(`[`, 2) #2.c
           ) |>
    mutate(
           condition =  gsub("ratio", "", sampleId) |> #3.a
               str_split("_") |> #3.b
               sapply(`[`, 1) |> #3.c
               as.numeric() |> #3.d
               as.factor(), #3.e
           rep = sampleId |> 
               str_split("_") |> #4.a
               sapply(`[`, 2) |> #4.b
               replace_na(replace = "1") |> #4.c
               as.factor(), #4.d
           ratio = condition |> 
             as.character() |> 
             as.double()
           )

2.3 Convert to QFeatures

First, recall that the precursor table is a file in long format. Every quantitative column in the precursor table contains information for multiple runs. Therefore, the function split the table based on the run identifier, given by the runCol argument (for Spectronaut, that identifier is contained in run). So, the QFeatures object after import will contain as many sets as there are runs. Next, the function links the annotation table with the PSM data. To achieve this, the annotation table must contain a variable named runCol that provides the run identifier in which each sample has been acquired, and this information will be used to match the identifiers in the Run column of the precursor table.

Here, we will use the FG_MS2RawQuantity column as quantification input.

Note, that we filter a number of variables to reduce the footprint of the QFeatures object.

(qf <- readQFeatures(assayData = precursors,  
                              colData = annot,
                              quantCols = "FG_MS2RawQuantity",
                              runCol = "R_FileName",
                              fnames = "EG_PrecursorId"))
## Checking arguments.
## Loading data as a 'SummarizedExperiment' object.
## Splitting data in runs.
## 
  |                                                                            
  |                                                                      |   0%
  |                                                                            
  |========                                                              |  11%
  |                                                                            
  |================                                                      |  22%
  |                                                                            
  |=======================                                               |  33%
  |                                                                            
  |===============================                                       |  44%
  |                                                                            
  |=======================================                               |  56%
  |                                                                            
  |===============================================                       |  67%
  |                                                                            
  |======================================================                |  78%
  |                                                                            
  |==============================================================        |  89%
  |                                                                            
  |======================================================================| 100%
## Formatting sample annotations (colData).
## Formatting data as a 'QFeatures' object.
## Setting assay rownames.
## An instance of class QFeatures (type: bulk) with 9 sets:
## 
##  [1] B000254_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio02_DIA: SummarizedExperiment with 44283 rows and 1 columns 
##  [2] B000258_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio04_DIA: SummarizedExperiment with 44283 rows and 1 columns 
##  [3] B000262_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio08_DIA: SummarizedExperiment with 44283 rows and 1 columns 
##  ...
##  [7] B000302_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio02_DIA_3: SummarizedExperiment with 44283 rows and 1 columns 
##  [8] B000306_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio04_DIA_3: SummarizedExperiment with 44283 rows and 1 columns 
##  [9] B000310_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio08_DIA_3: SummarizedExperiment with 44283 rows and 1 columns

3 Data preprocessing

The data preprocessing workflow for DIA data is similar to the workflow for DDA-LFQ data, but there are suble differences as we start from precursor level data and have additional columns.

3.1 Encoding missing values

We first replace any zero in the quantitative data with an NA.

qf <- zeroIsNA(qf, names(qf))

Note that msqrob2 can handle missing data without having to rely on hard-to-verify imputation assumptions, which is our general recommendation. However, msqrob2 does not prevent users from using imputation, which can be performed by starting from Spectronaut output with imputed values or by imputing the data themselves, e.g. by using impute() from the QFeatures package.

3.2 Precursor Filtering

Filtering removes low-quality and unreliable precursors that would otherwise introduce noise and artefacts in the data.

3.2.1 Remove questionable identifications

We apply standard filtering:

  1. q-value threshold of 0.01 for the identification of precursors (EG_Qvalue) and protein groups (PG_Qvalue).

  2. Remove precursors that could not be mapped, i.e. when EG_PrecursorId column is an empty string.

  3. Filter decoys, i.e. only keep precursors for which the EG_IsDecoy column equals 0.

  4. Keeping only proteotypic peptides, which map uniquely to a specific protein.

  5. Only keep non-imputed intensities: EG_IsImputed equals 0.

qf <-  qf |>
  filterFeatures(~ EG_Qvalue <= 0.01 & #1.
                     PG_Qvalue <= 0.01 &  #1.
#                     Lib.Q.Value <= 0.01 & #1.
#                     Lib.PG_Q.Value <= 0.01 & #1.
                     EG_PrecursorId != "" & #2.
                     EG_IsDecoy == 0 & #3.
                     PEP_IsProteotypic == 1 & #4
                     EG_IsImputed == 0) #5 
## 'EG_Qvalue' found in 9 out of 9 assay(s).'PG_Qvalue' found in 9 out of 9 assay(s).'EG_PrecursorId' found in 9 out of 9 assay(s).'EG_IsDecoy' found in 9 out of 9 assay(s).'PEP_IsProteotypic' found in 9 out of 9 assay(s).'EG_IsImputed' found in 9 out of 9 assay(s).

Note, that it is important that the filtering criteria are not distorting the distribution of the test statistics in the downstream analysis for features that are non-DA. It can be shown that filtering will not induce bias results when the filtering criterion is independent of test statistic. The criteria that we proposed above are all based on the results of the identification step, hence, they are independent of the downstream test statistics that will be used to prioritize DA proteins.

3.2.2 Assay joining

Up to now, the data from different runs were kept in separate assays. We can now join the filtered sets into an precursor set using joinAssays(). Sets are joined by stacking the columns (samples) in a matrix and rows (features) are matched according to a row identifier, here the EG_PrecursorId. We will store the result in the assay with name: precursor.

(qf <- joinAssays(
  x = qf,
  i = names(qf),
  fcol = "EG_PrecursorId",
  name = "precursors"
  ))
## Using 'EG_PrecursorId' to join assays.
## An instance of class QFeatures (type: bulk) with 10 sets:
## 
##  [1] B000254_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio02_DIA: SummarizedExperiment with 37045 rows and 1 columns 
##  [2] B000258_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio04_DIA: SummarizedExperiment with 37654 rows and 1 columns 
##  [3] B000262_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio08_DIA: SummarizedExperiment with 37184 rows and 1 columns 
##  ...
##  [8] B000306_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio04_DIA_3: SummarizedExperiment with 37323 rows and 1 columns 
##  [9] B000310_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio08_DIA_3: SummarizedExperiment with 36407 rows and 1 columns 
##  [10] precursors: SummarizedExperiment with 41857 rows and 9 columns

3.2.3 Filtering: Remove highly missing precursors

We keep peptides that were observed at last 4 times out of the \(n = 9\) samples, so that we can estimate the peptide characteristics. We tolerate the following proportion of NAs: \(\text{pNA} = \frac{(n - 4)}{n} = 0.556\), so we keep peptides that are observed in at least 44.4% of the samples, which corresponds to one treatment condition. This is an arbitrary value that may need to be adjusted depending on the experiment and the data set.

nObs <- 4
n <- ncol(qf[["precursors"]])

(qf <- filterNA(qf, i = "precursors", pNA = (n - nObs) / n))
## An instance of class QFeatures (type: bulk) with 10 sets:
## 
##  [1] B000254_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio02_DIA: SummarizedExperiment with 37045 rows and 1 columns 
##  [2] B000258_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio04_DIA: SummarizedExperiment with 37654 rows and 1 columns 
##  [3] B000262_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio08_DIA: SummarizedExperiment with 37184 rows and 1 columns 
##  ...
##  [8] B000306_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio04_DIA_3: SummarizedExperiment with 37323 rows and 1 columns 
##  [9] B000310_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio08_DIA_3: SummarizedExperiment with 36407 rows and 1 columns 
##  [10] precursors: SummarizedExperiment with 39126 rows and 9 columns

3.2.4 Filter one-hit wonders

Here, we remove proteins that can only be found by one peptide, as such proteins may not be trustworthy.

  1. We first calculate how many distinct peptides map to each protein group (PG_ProteinGroups). We use the stripped precursor sequence, i.e. sequence of the base peptide for this purpose.

  2. We store this information in the row data of the precursors assay

  3. We filter precursors of one-hit wonder proteins.

# Filter for peptides per protein
pepsPerProtDf <- qf[["precursors"]] |> 
    rowData() |> 
    data.frame() |>
    dplyr::select("PEP_StrippedSequence", "PG_ProteinGroups") |>
    group_by(PG_ProteinGroups) |>
    mutate(pepsPerProt = PEP_StrippedSequence |>
               unique() |> length()
           ) #1.

rowData(qf[["precursors"]])$pepsPerProt <- pepsPerProtDf$pepsPerProt #2. 

qf <- filterFeatures(qf, 
                              ~ pepsPerProt > 1, 
                              keep = TRUE) #3.
## 'pepsPerProt' found in 1 out of 10 assay(s).

3.3 Log-transformation

We perform log2-transformation with logTransform() from the QFeatures package. We use base = 2 and store the result in a new summarized experiment named precursors_log

qf <- logTransform(qf, 
                            base = 2, 
                            i = "precursors", 
                            name = "precursors_log")

3.4 Normalisation

The most common objective of MS-based proteomics experiments is to understand the biological changes in protein abundance between experimental conditions. However, changes in measurements between groups can be caused due to technical factors. For instance, there are systematic fluctuations from run-to-run that shift the measured intensity distribution. We can this explore as follows:

  1. We extract the sets containing the log transformed data. This is performed using QFeatures’ 3-way subsetting.
  2. We use longForm() to convert the QFeatures object into a long table, including condition and concentration for filtering and colouring.
  3. We visualise the density of the quantitative values within each sample. We colour each sample based on its spike-in condition.
qf[, , "precursors_log"] |> #1.
  longForm(colvars = c( "rep", "condition")) |>  #2. 
  data.frame() |> 
  filter(!is.na(value)) |>
  ggplot() + #3.
  aes(x = value,
      colour = condition,
      group = colname) +
  geom_density() +
  theme_minimal()
## Warning: 'experiments' dropped; see 'drops()'
## harmonizing input:
##   removing 18 sampleMap rows not in names(experiments)

Even in this clean synthetic data set (same yeast background with some spiked UPS proteins), the marginal precursor intensity distributions across samples are not well aligned. Ignoring this effect will increase the noise and reduce the statistical power of the experiment, and may also, in case of unbalanced designs, introduce confounding effects that will bias the results.

There are many ways to perform the normalisation, e.g.
median centering is a popular choice. If we subtract the sample median at the log scale from each precursor log2 intensity, this basically boils down to calculating log-ratio’s between the precursor intensity and its sample median. Here, we choose to work with offsets, that are chosen in such a way so that the distributions are centered around the location of a reference sample. E.g. the median of the sample medians.

Note that users can also use all normalisation functions that are provided in QFeatures, i.e. by replacing the sweep function by QFeatures::normalize.

qf <- sweep( #4. Subtract log2 norm factor column-by-column (MARGIN = 2)
  qf,
  MARGIN = 2,
  STATS = nfLogMedian(qf,"precursors_log"),
  i = "precursors_log",
  name = "precursors_norm"
)
## This function aims to calculate norm factors on a log scale, 
##           the input data are assumed to be on the log-scale!

We explore the effect of the global normalisation in the subsequent plot.

Formally, the function applies the following operation on each sample \(i\) across all precursors \(p\):

\[ y_{ip}^{\text{norm}} = y_{ip} - \log_2(nf_i) \]

with \(y_{ip}\) the log2-transformed intensities and \(\log_2(nf_i)\) the log2-transformed norm factor. Upon normalisation, we can see that the distribution of the \(y_{ip}^{\text{norm}}\) nicely overlap (using the same code as above)

qf[, , "precursors_norm"] |>
  longForm(colvars = colnames(colData(qf))) |> 
  data.frame() |> 
  filter(!is.na(value)) |>
  ggplot() + 
  aes(x = value,
      colour = condition,
      group = colname) +
  geom_density() +
  labs(subtitle = "Normalised log2 precursor intensities") +
  theme_minimal()
## Warning: 'experiments' dropped; see 'drops()'
## harmonizing input:
##   removing 27 sampleMap rows not in names(experiments)

3.5 Summarisation

Here, we summarise the precursor-level data into protein intensities using maxLFQ. maxLFQ first calculates all pairwise log ratio’s between samples only using their shared precursors. Particularly, it uses the median of the log ratio’s between the shared precursors, i.e. \[r_{ij} = median(y_{sj}-y_{si})\].
Hence, it first eliminates peptide effects. It then estimates the summaries by solving

\[ \sum_i\sum_j(y^{prot}_j - y^{prot}_i - r_ij)^2 \] It is implemented in the maxLFQ function of the iq package.

aggregateFeatures() streamlines summarisation. It requires the name of a rowData column to group the precursors into proteins (or protein groups), here PG_ProteinGroups. We provide the summarisation approach through the fun argument. Other summarisation methods are available from the MsCoreUtils package, see ?aggregateFeatures for a comprehensive list. The function will return a QFeatures object with a new set that we called proteins.

Note, that maxLFQ can take a long time for large datasets. In that case we suggest to replace function(X) iq::maxLFQ(X)$estimate by MsCoreUtils::medianPolish, na.rm = TRUE. Note that the optional argument na.rm = TRUE is needed when using medianPolish because the data contains missing values as we did not adopt imputation.

(qf <- aggregateFeatures(
  qf, i = "precursors_norm", 
  name = "proteins",
  fcol = "PG_ProteinGroups", 
  fun = function(X) iq::maxLFQ(X)$estimate
  #fun = MsCoreUtils::medianPolish, na.rm = TRUE
))
## 
  |                                                                            
  |                                                                      |   0%
  |                                                                            
  |======================================================================| 100%
## The following messages occurred during aggregation:
## 
## Your quantitative and row data contain missing values. Please read the
## relevant section(s) in the aggregateFeatures manual page regarding the
## effects of missing values on data aggregation.
## 
## Occurred during the aggregation of set(s): B000254_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio02_DIA
## An instance of class QFeatures (type: bulk) with 13 sets:
## 
##  [1] B000254_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio02_DIA: SummarizedExperiment with 37045 rows and 1 columns 
##  [2] B000258_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio04_DIA: SummarizedExperiment with 37654 rows and 1 columns 
##  [3] B000262_Ap_6883_EXT-765_DIA_Yeast_UPS2_ratio08_DIA: SummarizedExperiment with 37184 rows and 1 columns 
##  ...
##  [11] precursors_log: SummarizedExperiment with 38694 rows and 9 columns 
##  [12] precursors_norm: SummarizedExperiment with 38694 rows and 9 columns 
##  [13] proteins: SummarizedExperiment with 3143 rows and 9 columns

4 Data exploration and QC

Data exploration aims to highlight the main sources of variation in the data prior to data modelling and can pinpoint to outlying or off-behaving samples.

4.1 Marginal distribution at precursor and protein level

qf[, , "precursors_norm"] |>
  longForm(colvars = colnames(colData(qf))) |> 
  data.frame() |> 
  filter(!is.na(value)) |>
  ggplot() + 
  aes(x = value,
      colour = condition,
      group = colname) +
  geom_density() +
  theme_minimal() +
   labs(subtitle = "Normalised log2 precursor intensities")
## Warning: 'experiments' dropped; see 'drops()'
## harmonizing input:
##   removing 36 sampleMap rows not in names(experiments)

qf[, , "precursors_norm"] |>
  longForm(colvars = colnames(colData(qf))) |> 
  data.frame() |> 
  filter(!is.na(value)) |>
  ggplot() + 
  aes(x = sampleId,
      y = value,
      colour = condition,
      group = colname) +
  xlab("sample") +
  geom_boxplot() +
  theme_minimal() +
   labs(subtitle = "Normalised log2 precursor intensities")
## Warning: 'experiments' dropped; see 'drops()'
## harmonizing input:
##   removing 36 sampleMap rows not in names(experiments)

qf[, , "proteins"] |>
  longForm(colvars = colnames(colData(qf))) |> 
  data.frame() |> 
  filter(!is.na(value)) |>
  ggplot() + 
  aes(x = value,
      colour = condition,
      group = colname) +
  geom_density() +
  theme_minimal() +
   labs(subtitle = "Normalised log2 protein intensities")
## Warning: 'experiments' dropped; see 'drops()'
## harmonizing input:
##   removing 36 sampleMap rows not in names(experiments)

qf[, , "proteins"] |>
  longForm(colvars = colnames(colData(qf))) |> 
  data.frame() |> 
  filter(!is.na(value)) |>
  ggplot() + 
  aes(x = sampleId,
      y = value,
      colour = condition,
      group = colname) +
  xlab("sample") +
  geom_boxplot() +
  theme_minimal() +
   labs(subtitle = "Normalised log2 protein intensities")
## Warning: 'experiments' dropped; see 'drops()'
## harmonizing input:
##   removing 36 sampleMap rows not in names(experiments)

4.2 Charge state

qf[, , "precursors_norm"] |>
  longForm(colvars = colnames(colData(qf)), rowvars = "FG_Charge") |>
  as.data.frame() |>
  filter(!is.na(value)) |>
  filter(FG_Charge<=4) |> 
ggplot(aes(x = sampleId)) +
  geom_bar(aes(fill = factor(FG_Charge, levels = 4:1)),
           colour = "black") +
  labs(title = "Peptide types per sample",
       x = "Sample",
       fill = "Charge state") +
  theme_bw()
## Warning: 'experiments' dropped; see 'drops()'
## harmonizing input:
##   removing 36 sampleMap rows not in names(experiments)

4.3 Identifications per sample

qf[,,"precursors_norm"] |> 
  longForm(colvars = colnames(colData(qf)), 
           rowvars= c("EG_PrecursorId", 
                      "PG_ProteinGroups", 
                      "PEP_StrippedSequence",
                      "FG_Charge")) |>
  data.frame() |>
  filter(!is.na(value)) |>
  mutate(cond_rep = paste0("ratio", condition, "_", rep)) |>
  group_by(condition, cond_rep) |>
  summarise(Precursors = length(unique(EG_PrecursorId)),
            `Protein Groups` = length(unique(PG_ProteinGroups))) |> 
  pivot_longer(-(1:2),
               names_to = "Feature",
               values_to = "IDs") |> 
  ggplot(aes(x = cond_rep, y = IDs, fill = condition)) +
  geom_col() +
  #scale_fill_observable() +
  facet_wrap(~Feature,
             scales = "free_y") +
  labs(title = "Precursor and protein group identificiations per sample",
       x = "Sample",
       y = "Identifications") +
  theme_bw() + 
  theme(axis.text.x = element_text(angle = 90))
## Warning: 'experiments' dropped; see 'drops()'
## harmonizing input:
##   removing 36 sampleMap rows not in names(experiments)
## `summarise()` has regrouped the output.
## ℹ Summaries were computed grouped by condition and cond_rep.
## ℹ Output is grouped by condition.
## ℹ Use `summarise(.groups = "drop_last")` to silence this message.
## ℹ Use `summarise(.by = c(condition, cond_rep))` for per-operation grouping
##   (`?dplyr::dplyr_by`) instead.

4.4 Dimensionality reduction plot

A common approach for data exploration is to perform dimension reduction, such as Multi Dimensional Scaling (MDS). We will first extract the set to explore along the sample annotations (used for plot colouring).

getWithColData(qf, "proteins") |> 
  as("SingleCellExperiment") |> 
  runMDS(exprs_values = 1) |> 
  plotMDS(colour_by = "condition", shape_by = "rep", point_size = 3) 
## Warning: 'experiments' dropped; see 'drops()'

This plot reveals interesting information. First, we see that the samples are nicely separated according to their spike-in condition. Interestingly, the conditions are sorted by the concentration.

4.5 Correlation matrix

corMat <- qf[["proteins"]] |> 
  assay() |>
  cor(method = "spearman", use = "pairwise.complete.obs") 
 
colnames(corMat) <- qf$sampleId
rownames(corMat) <- qf$sampleId
corMat |> 
  ggcorrplot::ggcorrplot() +
  scale_fill_viridis_c() +
  labs(title = "Correlation matrix",
       fill = "Correlation") +
  theme(axis.text.x = element_text(angle = 90))
## Scale for fill is already present.
## Adding another scale for fill, which will replace the existing scale.

5 Data Modeling (Robust Regression)

5.1 Sources of variation

For this experiment, we can only model a single source of variation: spike-in concentration. All remaining sources of variability, e.g. pipetting errors, MS run to run variability, etc will be lumped in the residual variability (error term).

Spike-in condition effects: we model the source of variation induced by the experimental treatment of interest as a fixed effect, which we consider non-random, i.e. the treatment effect is assumed to be the same in repeated experiments, but it is unknown and has to be estimated.

When modelling a typical label-free experiment at the protein level, the model boils down to the following linear model:

\[ y_i = \mathbf{x}_i^T\boldsymbol{\beta} +\epsilon_i\] with the \(\log_2\)-normalised and summarised protein intensities in sample; \(\mathbf{x}_i\) a vector with the covariate pattern for the sample encoding the intercept, spike-in condition, or other experimental factors;
\(\boldsymbol{\beta}\) the vector of parameters that model the association between the covariates and the outcome; and \(\epsilon_i\) the residuals reflecting variation that is not captured by the fixed effects. Note that allows for a flexible parameterisation of the treatment beyond a single covariate, i.e. including a 1 for the intercept, continuous and categorical variables as well as their interactions. We assume the residuals to be independent and identically distributed (i.i.d) according to a normal distribution with zero mean and constant variance, i.e. \(\epsilon_i \sim N(0, \sigma^2_\epsilon)\), that can differ from protein to protein.

In R, this model is encoded using the following simple formula:

model <- ~ condition

5.2 Model Estimation

We estimate the model with msqrob(). The function takes the QFeatures object, extracts the quantitative values from the "proteins" set generated during summarisation, and fits a simple linear model with condition as covariate, which are automatically retrieved from colData(qf). Note, that msqrob() can also take a SummarizedExperiment of precursor, peptide or protein intensities as its input.

  qf <- msqrob(
    qf,  
    i = "proteins",
    formula = model,
    robust = TRUE)

We enable M-estimation (robust = TRUE) for improved robustness against outliers.

The fitting results are available in the msqrobModels column of the rowData. More specifically, the modelling output is stored in the rowData as a statModel object, one model per row (protein). We will see in a later section how to perform statistical inference on the estimated parameters.

models <- rowData(qf[["proteins"]])[["msqrobModels"]]
models[1:3]
## $D6VTK4
## Object of class "StatModel"
## The type is "rlm"
## There number of elements is 5 
## 
## $O13297
## Object of class "StatModel"
## The type is "rlm"
## There number of elements is 5 
## 
## $O13525
## Object of class "StatModel"
## The type is "rlm"
## There number of elements is 5

5.3 Statistical inference

We can now convert the research question “do the spike-in conditions affect the protein intensities?” into a statistical hypothesis.

In other words, we need to translate this question in a null and alternative hypothesis on a single model parameter or on a linear combination of model parameters, which is also referred to with a contrast.

To aid defining contrasts, we will visualise the experimental design using the ExploreModelMatrix package.

vd <- ExploreModelMatrix::VisualizeDesign(
    sampleData =  colData(qf),
    designFormula = model,
    textSizeFitted = 4
)
vd$plotlist
## [[1]]

This plot shows that the average protein \(\log_2\) intensity for condition2 is modelled by (Intercept), for the condition4 group by (Intercept) + condtion4 and for the condition8 group by (Intercept) + condtion8.

Hence, to assess differential abundance of proteins between spike-in conditions, we will have to assess all pairwise combinations between the spike-in groups, which leads to 3 different contrasts, i.e. the log2 fold change between ratio 4 and ratio 2, \(\log_2 FC_{4-2} = (Intercept + condition4) - Intercept = condition4\), the log2 fold change between ratio 8 and ratio 2, \(\log_2 FC_{8-2} = (Intercept + condition8) - Intercept = condition8\), and the log2 fold change between ratio 8 and ratio 4, \(\log_2 FC_{8-4} = (Intercept + condition8) - (Intercept+condition4) = condition8 - condition4\).

We can generate the null-hypothesess for all pairwise comparisons manually. However, here we will use the createPairwiseContrasts function in msqrob2.

(allHypotheses <- createPairwiseContrasts(
  model, colData(qf), "condition"
  )
)
## [1] "condition4 = 0"              "condition8 = 0"             
## [3] "condition8 - condition4 = 0"

Based on these null hypothesis we now generate the contrast matrix with all three contrasts using the makeContrast function. Note, that experienced users can also define the contrast matrix, directly.

(L <- makeContrast(
  allHypotheses,
  parameterNames = colnames(vd$designmatrix)
))
##             condition4 condition8 condition8 - condition4
## (Intercept)          0          0                       0
## condition4           1          0                      -1
## condition8           0          1                       1

We can now test our null hypothesis using hypothesisTest() which takes the QFeatures object with the fitted model and the contrast we just built. msqrob2 automatically applies the hypothesis testing to all features in the assay.

qf <- hypothesisTest(qf, i = "proteins", contrast = L)

The results tables for each contrast are stored in the row data of the proteins assay in qf under the colomnames of the contrast matrix L.

qf[["proteins"]] |> 
  rowData() |> 
  names()
##  [1] "PG_Genes"                                     
##  [2] "PG_ProteinAccessions"                         
##  [3] "PG_ProteinGroups"                             
##  [4] "PG_ProteinNames"                              
##  [5] "PG_Organisms"                                 
##  [6] "PG_NrOfPrecursorsIdentified_.Experiment.wide."
##  [7] "PG_Qvalue"                                    
##  [8] "EG_IsDecoy"                                   
##  [9] "PEP_IsProteotypic"                            
## [10] "EG_IsImputed"                                 
## [11] "FG_PrecursorSignalToNoise"                    
## [12] "species"                                      
## [13] "pepsPerProt"                                  
## [14] ".n"                                           
## [15] "msqrobModels"                                 
## [16] "condition4"                                   
## [17] "condition8"                                   
## [18] "condition8 - condition4"

Users can directly access the results tables via de colData. However, msqrob2 also provides the msqrobCollect function. This function will automatically retrieve the results tables for all contrasts and combine them by default in one table. This table contains two additional columns: contrast with the name of the contrast and feature with the nme of the modelled feature, here protein group.

inferences <- 
  msqrobCollect(qf[["proteins"]], L) 
head(inferences)
##                         logFC         se       df         t        pval
## condition4.D6VTK4 -0.95827900 0.22295001 7.173801 -4.298179 0.003376991
## condition4.O13297  0.09932233 0.06643463 7.898270  1.495039 0.173744355
## condition4.O13525  0.44999153 0.43101711 7.763115  1.044022 0.327891822
## condition4.O13539  0.25487895 0.20159528 7.252666  1.264310 0.245232762
## condition4.O13563  0.20329728 0.12402293 8.000403  1.639191 0.139803508
## condition4.O13585  0.29012155 0.15953759 7.692916  1.818515 0.107981907
##                     adjPval   contrast feature
## condition4.D6VTK4 0.2527115 condition4  D6VTK4
## condition4.O13297 0.9465407 condition4  O13297
## condition4.O13525 0.9909466 condition4  O13525
## condition4.O13539 0.9624588 condition4  O13539
## condition4.O13563 0.9388941 condition4  O13563
## condition4.O13585 0.8870638 condition4  O13585

Note, that missing values can occur because data modelling resulted in a fitError. We refer users to our section 2.9 of our msqrob2 book where we elaborate on how to deal with proteins that could not be fit.

The results also include an adjusted p-value for each protein to correct for multiple testing using the Benjamini-Hochberg False Discovery Rate (FDR) method, which represents an estimate of the minimum FDR at which the test result for protein is considered significant, i.e. the expected fraction of false positives that are reported in the shortest top list of the most significant proteins that include protein \(j\).

Note, that for this design we exceptionally also know the ground truth. So we will also add this information to the inference table:

inferences <- inferences |> 
    mutate(species =  rep(rowData(qf[["proteins"]])$species, ncol(L)))

5.3.1 Results tables for significant proteins

We will return tables of proteins for which the contrasts are significant at the 5% FDR level.

  1. We set the significance level
  2. We loop over the contrasts
  3. We filter the significant results for the contrast from the table
  4. We print the table
alpha <- 0.05 #1.
for (contrasti in colnames(L)) #2. 
{
  sigList <- inferences |> 
    filter(contrast == contrasti & adjPval < alpha) #3.
  cat("**Contrast:**", contrasti, "= 0 (", nrow(sigList), " significant proteins)\n\n") #4.
  print(kable(sigList) |>
      kable_styling(full_width = FALSE) |>
      scroll_box(height = "250px")
      ) #4.
  cat("\n\n\n---\n\n") #4.
}

Contrast: condition4 = 0 ( 22 significant proteins)

logFC se df t pval adjPval contrast feature species
condition4.O76070ups 1.4165909 0.1366644 7.220883 10.365468 0.0000136 0.0029735 condition4 O76070ups ups
condition4.P00167ups 0.8530018 0.1143329 7.827509 7.460686 0.0000805 0.0126541 condition4 P00167ups ups
condition4.P00915ups 0.9817079 0.0938298 7.663583 10.462650 0.0000083 0.0023735 condition4 P00915ups ups
condition4.P00918ups 1.0294513 0.0979739 7.083382 10.507406 0.0000142 0.0029735 condition4 P00918ups ups
condition4.P01008ups 1.4680568 0.1189790 7.847467 12.338791 0.0000020 0.0014271 condition4 P01008ups ups
condition4.P02144ups 1.0073663 0.1016046 7.866129 9.914576 0.0000102 0.0026698 condition4 P02144ups ups
condition4.P02768ups 1.0320544 0.0823880 7.032368 12.526754 0.0000046 0.0018024 condition4 P02768ups ups
condition4.P04040ups 0.9353175 0.0867016 7.842685 10.787778 0.0000056 0.0018113 condition4 P04040ups ups
condition4.P06732ups 1.1730488 0.0702199 7.399871 16.705367 0.0000004 0.0006005 condition4 P06732ups ups
condition4.P12081ups 1.0787800 0.0711228 8.000403 15.167853 0.0000004 0.0006005 condition4 P12081ups ups
condition4.P15559ups 1.0978851 0.1025832 7.872578 10.702386 0.0000058 0.0018113 condition4 P15559ups ups
condition4.P25588 0.6382375 0.0844067 7.840964 7.561457 0.0000727 0.0120221 condition4 P25588 yeast
condition4.P28003 -0.7901346 0.1265320 7.618709 -6.244544 0.0003024 0.0431954 condition4 P28003 yeast
condition4.P55957ups 1.6694380 0.1630636 7.272334 10.237958 0.0000141 0.0029735 condition4 P55957ups ups
condition4.P62937ups 1.0348592 0.0839790 7.594775 12.322828 0.0000027 0.0014271 condition4 P62937ups ups
condition4.P62988ups 1.0937870 0.1199312 6.364060 9.120124 0.0000694 0.0120221 condition4 P62988ups ups
condition4.P63165ups 0.9845836 0.0707820 7.797938 13.910090 0.0000009 0.0009212 condition4 P63165ups ups
condition4.P63279ups 0.8720731 0.1255320 7.554782 6.947021 0.0001557 0.0233082 condition4 P63279ups ups
condition4.P68871ups 0.8838205 0.0979298 7.820805 9.025045 0.0000210 0.0038814 condition4 P68871ups ups
condition4.P69905ups 1.0502355 0.0822797 7.529061 12.764218 0.0000023 0.0014271 condition4 P69905ups ups
condition4.Q06830ups 0.8761396 0.0780639 7.886837 11.223365 0.0000040 0.0017914 condition4 Q06830ups ups
condition4.Q15843ups 1.5161987 0.1621133 7.718274 9.352711 0.0000177 0.0034773 condition4 Q15843ups ups

Contrast: condition8 = 0 ( 27 significant proteins)

logFC se df t pval adjPval contrast feature species
condition8.O76070ups 2.802893 0.1366644 7.220883 20.509304 1.00e-07 0.0000226 condition8 O76070ups ups
condition8.P00167ups 2.045412 0.1126240 7.827509 18.161427 1.00e-07 0.0000226 condition8 P00167ups ups
condition8.P00915ups 2.070374 0.0967510 7.663583 21.398987 0.00e+00 0.0000099 condition8 P00915ups ups
condition8.P00918ups 2.282998 0.0981482 7.083382 23.260724 1.00e-07 0.0000134 condition8 P00918ups ups
condition8.P01008ups 3.190077 0.1189790 7.847467 26.812103 0.00e+00 0.0000040 condition8 P01008ups ups
condition8.P01031ups 2.035147 0.1996097 7.691241 10.195630 9.80e-06 0.0012261 condition8 P01031ups ups
condition8.P02144ups 2.161638 0.1027879 7.866129 21.030091 0.00e+00 0.0000089 condition8 P02144ups ups
condition8.P02753ups 2.423928 0.2127367 7.562368 11.394030 5.00e-06 0.0007101 condition8 P02753ups ups
condition8.P02768ups 2.185605 0.0775259 7.032368 28.191938 0.00e+00 0.0000060 condition8 P02768ups ups
condition8.P04040ups 2.136115 0.0855233 7.842685 24.977000 0.00e+00 0.0000041 condition8 P04040ups ups
condition8.P06732ups 2.274834 0.0702199 7.399871 32.395862 0.00e+00 0.0000035 condition8 P06732ups ups
condition8.P08263ups 2.220606 0.2356691 7.453054 9.422559 2.11e-05 0.0024588 condition8 P08263ups ups
condition8.P12081ups 2.235157 0.0711228 8.000403 31.426734 0.00e+00 0.0000035 condition8 P12081ups ups
condition8.P15559ups 2.238270 0.1025832 7.872578 21.819065 0.00e+00 0.0000079 condition8 P15559ups ups
condition8.P16083ups 2.047075 0.1436145 7.696505 14.253959 8.00e-07 0.0001240 condition8 P16083ups ups
condition8.P41159ups 2.028823 0.1693572 6.661218 11.979548 9.40e-06 0.0012261 condition8 P41159ups ups
condition8.P53733 2.121012 0.2339276 8.000403 9.066960 1.75e-05 0.0021206 condition8 P53733 yeast
condition8.P55957ups 2.919539 0.1630636 7.272334 17.904300 3.00e-07 0.0000487 condition8 P55957ups ups
condition8.P61626ups 2.033475 0.1915112 8.000403 10.618047 5.40e-06 0.0007397 condition8 P61626ups ups
condition8.P62937ups 2.158332 0.0828064 7.594775 26.064779 0.00e+00 0.0000041 condition8 P62937ups ups
condition8.P62988ups 2.176507 0.1157211 6.364060 18.808220 8.00e-07 0.0001240 condition8 P62988ups ups
condition8.P63165ups 2.042465 0.0707820 7.797938 28.855718 0.00e+00 0.0000035 condition8 P63165ups ups
condition8.P63279ups 2.270314 0.1250760 7.554782 18.151482 2.00e-07 0.0000308 condition8 P63279ups ups
condition8.P68871ups 2.159318 0.0994765 7.820805 21.706808 0.00e+00 0.0000082 condition8 P68871ups ups
condition8.P69905ups 2.160724 0.0786942 7.529061 27.457224 0.00e+00 0.0000041 condition8 P69905ups ups
condition8.Q06830ups 2.026019 0.0780639 7.886837 25.953349 0.00e+00 0.0000040 condition8 Q06830ups ups
condition8.Q15843ups 2.519950 0.1621133 7.718274 15.544378 4.00e-07 0.0000698 condition8 Q15843ups ups

Contrast: condition8 - condition4 = 0 ( 22 significant proteins)

logFC se df t pval adjPval contrast feature species
condition8 - condition4.O76070ups 1.386302 0.1260488 7.220883 10.998134 0.0000091 0.0016753 condition8 - condition4 O76070ups ups
condition8 - condition4.P00167ups 1.192410 0.1143329 7.827509 10.429281 0.0000073 0.0014302 condition8 - condition4 P00167ups ups
condition8 - condition4.P00915ups 1.088666 0.0967510 7.663583 11.252242 0.0000049 0.0010276 condition8 - condition4 P00915ups ups
condition8 - condition4.P00918ups 1.253546 0.0892090 7.083382 14.051793 0.0000020 0.0006196 condition8 - condition4 P00918ups ups
condition8 - condition4.P01008ups 1.722020 0.1174127 7.847467 14.666390 0.0000006 0.0003486 condition8 - condition4 P01008ups ups
condition8 - condition4.P02144ups 1.154272 0.1027879 7.866129 11.229650 0.0000041 0.0009807 condition8 - condition4 P02144ups ups
condition8 - condition4.P02768ups 1.153551 0.0836655 7.032368 13.787650 0.0000024 0.0006837 condition8 - condition4 P02768ups ups
condition8 - condition4.P04040ups 1.200798 0.0867016 7.842685 13.849778 0.0000009 0.0004511 condition8 - condition4 P04040ups ups
condition8 - condition4.P06732ups 1.101785 0.0661999 7.399871 16.643308 0.0000004 0.0003486 condition8 - condition4 P06732ups ups
condition8 - condition4.P12081ups 1.156377 0.0711228 8.000403 16.258882 0.0000002 0.0003486 condition8 - condition4 P12081ups ups
condition8 - condition4.P15559ups 1.140385 0.1014606 7.872578 11.239685 0.0000040 0.0009807 condition8 - condition4 P15559ups ups
condition8 - condition4.P16083ups 1.251298 0.1476060 7.696505 8.477287 0.0000362 0.0059844 condition8 - condition4 P16083ups ups
condition8 - condition4.P39702 -1.037604 0.1595180 8.000403 -6.504621 0.0001871 0.0280100 condition8 - condition4 P39702 yeast
condition8 - condition4.P55957ups 1.250101 0.1513856 7.272334 8.257725 0.0000601 0.0094428 condition8 - condition4 P55957ups ups
condition8 - condition4.P62937ups 1.123473 0.0822817 7.594775 13.653980 0.0000013 0.0005514 condition8 - condition4 P62937ups ups
condition8 - condition4.P62988ups 1.082720 0.1058912 6.364060 10.224839 0.0000349 0.0059844 condition8 - condition4 P62988ups ups
condition8 - condition4.P63165ups 1.057881 0.0695350 7.797938 15.213639 0.0000004 0.0003486 condition8 - condition4 P63165ups ups
condition8 - condition4.P63279ups 1.398241 0.1210098 7.554782 11.554778 0.0000045 0.0010169 condition8 - condition4 P63279ups ups
condition8 - condition4.P68871ups 1.275497 0.0994765 7.820805 12.822094 0.0000016 0.0005514 condition8 - condition4 P68871ups ups
condition8 - condition4.P69905ups 1.110489 0.0822797 7.529061 13.496515 0.0000015 0.0005514 condition8 - condition4 P69905ups ups
condition8 - condition4.Q06830ups 1.149880 0.0773072 7.886837 14.874164 0.0000005 0.0003486 condition8 - condition4 Q06830ups ups
condition8 - condition4.Q15843ups 1.003751 0.1580632 7.718274 6.350316 0.0002566 0.0366624 condition8 - condition4 Q15843ups ups

5.3.2 Volcanoplots

A volcano plot is a common visualisation that provides an overview of the hypothesis testing results, plotting the \(-\log_{10}\) p-value1 Note, that upon this transformation a value of 1 represents a p-value of 0.1, 2 a p-value of 0.01, 3 a p-value of 0.001, etc. as a function of the estimated log fold change. Volcano plots can be used to highlight the most interesting proteins that have large fold changes and/or are highly significant. We can use the table above directly to build a volcano plot using ggplot2 functionality. We also highlight which proteins are UPS standards, known to be differentially abundant by experimental design.

Experienced users can make the plot themselves. However, in msqrob2 we also provide the plotVolcano function that generates volcanoplots based on msqrob2 inference tables generated with the hypothesisTest function.

Since the inference table contains multiple contrast we use the facet_wrap function to make a separate volcano plot for each contrast.

volcanoplots <- inferences |> 
  plotVolcano() +
  facet_wrap(~contrast)
volcanoplots

5.3.3 Heatmaps

We construct heatmaps for the significant proteins for each contrast.

  1. We will loop over the column names of the contrast matrix L with a user-defined function that extracts
  2. the names of the significant features for a contrast
  3. the quant data for the significant proteins and scale it
  4. annotation,
  5. and returns the heatmap using the quants data. We do not cluster columns (samples) to keep them together according to the design.
  6. We feed the function with the assay proteins along with its colData and
  7. the significance level alpha
heatmaps <- lapply(
  colnames(L), #1.
  function(contrast, se, alpha)
    {
    sig <- rowData(se)[[contrast]] |> #2.
      filter(adjPval < alpha) |> 
      rownames()
    
    quants <- t(scale(t(assay(se[sig,])))) #3
    
    colnames(quants) <- paste0("con", se$condition,"_rep",se$rep) #4.
    annotations <- columnAnnotation(condition = se$condition) #4.
    set.seed(1234) ## annotation colours are randomly generated by default
    return( #.5
                       Heatmap(
                       quants, 
                       name = "log2 intensity",
                       top_annotation = annotations, 
                       column_title = paste0(contrast, " = 0")
                       )
                     )
                   },
                   se = getWithColData(qf, "proteins"), #6.
                   alpha = alpha) #7.
## Warning: 'experiments' dropped; see 'drops()'
heatmaps
## [[1]]

## 
## [[2]]

## 
## [[3]]

A typical workflow for the analysis of a proteomics experiment will end.

6 Assess performance

Note, that the evaluations in this section are typically not possible on real experimental data. Indeed, we are using a spike-in dataset so we know the ground truth: all human UPS proteins are differentially abundant (DA) between the conditions and the yeast proteins are non-DA.

We use a nominal FDR level of 0.05

alpha <- 0.05

6.1 Real Fold changes

As this is a spike-in study with known ground truth, we can also plot the log2 fold change distributions against the expected values, in this case 0 for the yeast proteins and the difference of the log concentration for the spiked-in UPS standards. We first create a small table with the real values.

  1. We extract the levels from factor condition
  2. We convert then into a number as they refer to the real ratio
  3. We log2-transform them
  4. We calculate all pairwise differences between log2 transformed ratio’s to obtain the log2 fold changes

We use a similar operation to construct the name of the corresponding contrast.

(realLogFC <- data.frame(
    logFC = colData(qf)$condition |> 
        levels() |> # 1. 
        as.double() |> # 2. 
        log2() |> # 3. 
        combn(m=2, FUN = diff), #4.
    contrast = colData(qf)$condition |> 
        levels() |> 
        combn(m=2, FUN= function(x) paste0(x[2]," - ",x[1])) |>
        recode("4 - 2" = "condition4", 
               "8 - 2" = "condition8",
               "8 - 4" = "condition8 - condition4")
    )
)
##   logFC                contrast
## 1     1              condition4
## 2     2              condition8
## 3     1 condition8 - condition4

We now evaluate if the estimated fold changes correspond to the real fold changes.

  1. We plot the log fold changes in function of the spikein condition and color according to spikein condition.
  2. We add a boxplot layer.
  3. We use custom colors.
  4. We add a horizontal line at 0, which corresponds to the known log2 fold change for yeast proteins
  5. We add a horizontal lines for known log2 fold changes of the spiked UPS proteins.
logFC <- inferences |> 
  filter(!is.na(logFC)) |>
  ggplot(aes(x = species, y = logFC, color= species)) + #1.
  geom_boxplot() + #2.
  theme_bw() + 
  scale_color_manual( 
    values = c("grey20", "firebrick"), #3. 
    name = "",
    labels = c("yeast", "ups")
    ) + 
  geom_hline(yintercept = 0, color="grey20") + # 4. 
  facet_wrap(~contrast) +
  geom_hline(aes(yintercept = logFC), color="firebrick", data=realLogFC) #5. 
logFC

6.2 True and false positives

All human UPS proteins are differentially abundant (DA) between the conditions and the yeast proteins are non-DA.

tpFpTable <- group_by(inferences, contrast) |>
    filter(adjPval < alpha) |>
    summarise("TP" = sum(species == "ups"),
              "FP" = sum(species != "ups"),
              "FDP" = mean(species != "ups")) 
tpFpTable
## # A tibble: 3 × 4
##   contrast                   TP    FP    FDP
##   <chr>                   <int> <int>  <dbl>
## 1 condition4                 20     2 0.0909
## 2 condition8                 26     1 0.0370
## 3 condition8 - condition4    21     1 0.0455

6.3 TPR - FDP curves

We generate the TPR-FDP curves to assess the performance to prioritise differentially abundant proteins. Again, these curves are built using the ground truth information about which proteins are differentially abundant (spiked in) and which proteins are constant across samples. We create two functions to compute the TPR and the FDP.

computeFDP <- function(pval, tp) {
    ord <- order(pval)
    fdp <- cumsum(!tp[ord]) / 1:length(tp)
    fdp[order(ord)]
}
computeTPR <- function(pval, tp, nTP = NULL) {
    if (is.null(nTP)) nTP <- sum(tp)
    ord <- order(pval)
    tpr <- cumsum(tp[ord]) / nTP
    tpr[order(ord)]
}

We apply these functions and compute the corresponding metric using the statistical inference results and the ground truth information.

performance <- inferences |> group_by(contrast) |> 
    na.exclude() |>
    mutate(tpr = computeTPR(pval, species == "ups"),
           fdp = computeFDP(pval, species == "ups")) |>
    arrange(fdp)

We also highlight the observed FDP at a \(alpha = 5\%\) FDR threshold.

workPoints <- performance |> 
  group_by(contrast) |>
    filter(adjPval < 0.05) |>
    slice_max(pval)
ggplot(performance) +
    aes(
        y = fdp,
        x = tpr,
    ) +
    geom_line() +
    geom_point(data = workPoints, size = 3) +
    geom_hline(yintercept = 0.05, linetype = 2) +
    facet_wrap( ~ contrast) +
    coord_flip(ylim = c(0, 0.2)) +
    theme_minimal()

6.4 Volcano-plots

We make a separate plot per contrast and indicate the ground truth in the plot.

  inferences |> 
  plotVolcano() + 
  facet_wrap(~contrast) +
  aes(color = species, shape = adjPval < alpha) + 
    scale_color_manual(
      values = c("grey20", "firebrick"), 
      name = "species",
      labels = c("yeast", "ups")
      ) +
  labs(shape="FDR < 0.05") +
  geom_vline(aes(xintercept = logFC), data = realLogFC , col ="firebrick") +
  geom_hline(aes(yintercept = -log10(adjAlpha)), data = inferences |> 
               group_by(contrast) |> 
               summarize(adjAlpha = alpha *mean(adjPval < alpha, na.rm=TRUE), .groups="drop"))
## Scale for colour is already present.
## Adding another scale for colour, which will replace the existing scale.

7 References

Staes, An, Teresa Mendes Maia, Sara Dufour, Robbin Bouwmeester, Ralf Gabriels, Lennart Martens, Kris Gevaert, Francis Impens, and Simon Devos. 2024. “Benefit of in Silico Predicted Spectral Libraries in Data‑Independent Acquisition Data Analysis Workflows.” Journal of Proteome Research 23 (6): 2078–89. https://doi.org/10.1021/acs.jproteome.4c00048.

Appendix

Note, that more examples can be found in our msqrob2 book.

sessionInfo()
## R version 4.6.0 alpha (2026-04-05 r89794)
## Platform: x86_64-pc-linux-gnu
## Running under: Ubuntu 24.04.4 LTS
## 
## Matrix products: default
## BLAS:   /home/biocbuild/bbs-3.23-bioc/R/lib/libRblas.so 
## LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.12.0  LAPACK version 3.12.0
## 
## locale:
##  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
##  [3] LC_TIME=en_GB              LC_COLLATE=C              
##  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
##  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
##  [9] LC_ADDRESS=C               LC_TELEPHONE=C            
## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       
## 
## time zone: America/New_York
## tzcode source: system (glibc)
## 
## attached base packages:
## [1] grid      stats4    stats     graphics  grDevices utils     datasets 
## [8] methods   base     
## 
## other attached packages:
##  [1] scater_1.39.4               scuttle_1.21.6             
##  [3] SingleCellExperiment_1.33.2 ComplexHeatmap_2.27.1      
##  [5] kableExtra_1.4.0            ExploreModelMatrix_1.23.1  
##  [7] stringr_1.6.0               msqrob2_1.19.1             
##  [9] ggplot2_4.0.2               tidyr_1.3.2                
## [11] dplyr_1.2.1                 QFeatures_1.21.2           
## [13] MultiAssayExperiment_1.37.4 SummarizedExperiment_1.41.1
## [15] Biobase_2.71.0              GenomicRanges_1.63.2       
## [17] Seqinfo_1.1.0               IRanges_2.45.0             
## [19] S4Vectors_0.49.1            BiocGenerics_0.57.0        
## [21] generics_0.1.4              MatrixGenerics_1.23.0      
## [23] matrixStats_1.5.0           BiocStyle_2.39.0           
## 
## loaded via a namespace (and not attached):
##   [1] RColorBrewer_1.1-3      rstudioapi_0.18.0       jsonlite_2.0.0         
##   [4] shape_1.4.6.1           magrittr_2.0.5          magick_2.9.1           
##   [7] ggbeeswarm_0.7.3        farver_2.1.2            nloptr_2.2.1           
##  [10] rmarkdown_2.31          GlobalOptions_0.1.4     vctrs_0.7.3            
##  [13] Cairo_1.7-0             minqa_1.2.8             tinytex_0.59           
##  [16] BiocBaseUtils_1.13.0    htmltools_0.5.9         S4Arrays_1.11.1        
##  [19] BiocNeighbors_2.5.4     SparseArray_1.11.13     sass_0.4.10            
##  [22] bslib_0.10.0            htmlwidgets_1.6.4       plyr_1.8.9             
##  [25] cachem_1.1.0            igraph_2.2.3            mime_0.13              
##  [28] lifecycle_1.0.5         iterators_1.0.14        pkgconfig_2.0.3        
##  [31] rsvd_1.0.5              Matrix_1.7-5            R6_2.6.1               
##  [34] fastmap_1.2.0           rbibutils_2.4.1         shiny_1.13.0           
##  [37] clue_0.3-68             digest_0.6.39           colorspace_2.1-2       
##  [40] irlba_2.3.7             textshaping_1.0.5       beachmat_2.27.5        
##  [43] labeling_0.4.3          abind_1.4-8             compiler_4.6.0         
##  [46] bit64_4.6.0-1           withr_3.0.2             doParallel_1.0.17      
##  [49] S7_0.2.1                ggcorrplot_0.1.4.1      BiocParallel_1.45.0    
##  [52] viridis_0.6.5           iq_2.0.1                MASS_7.3-65            
##  [55] DelayedArray_0.37.1     rjson_0.2.23            tools_4.6.0            
##  [58] vipor_0.4.7             otel_0.2.0              beeswarm_0.4.0         
##  [61] httpuv_1.6.17           glue_1.8.0              nlme_3.1-169           
##  [64] promises_1.5.0          cluster_2.1.8.2         reshape2_1.4.5         
##  [67] gtable_0.3.6            data.table_1.18.2.1     utf8_1.2.6             
##  [70] BiocSingular_1.27.1     ScaledMatrix_1.19.0     xml2_1.5.2             
##  [73] XVector_0.51.0          ggrepel_0.9.8           foreach_1.5.2          
##  [76] pillar_1.11.1           limma_3.67.1            later_1.4.8            
##  [79] rintrojs_0.3.4          circlize_0.4.18         splines_4.6.0          
##  [82] lattice_0.22-9          bit_4.6.0               tidyselect_1.2.1       
##  [85] knitr_1.51              gridExtra_2.3           reformulas_0.4.4       
##  [88] bookdown_0.46           ProtGenerics_1.43.0     svglite_2.2.2          
##  [91] xfun_0.57               shinydashboard_0.7.3    statmod_1.5.1          
##  [94] DT_0.34.0               stringi_1.8.7           lazyeval_0.2.3         
##  [97] yaml_2.3.12             boot_1.3-32             evaluate_1.0.5         
## [100] codetools_0.2-20        MsCoreUtils_1.23.9      tibble_3.3.1           
## [103] BiocManager_1.30.27     cli_3.6.6               arrow_23.0.1.2         
## [106] xtable_1.8-8            systemfonts_1.3.2       Rdpack_2.6.6           
## [109] jquerylib_0.1.4         dichromat_2.0-0.1       Rcpp_1.1.1             
## [112] png_0.1-9               parallel_4.6.0          assertthat_0.2.1       
## [115] AnnotationFilter_1.35.0 lme4_2.0-1              viridisLite_0.4.3      
## [118] scales_1.4.0            purrr_1.2.2             crayon_1.5.3           
## [121] GetoptLong_1.1.1        rlang_1.2.0             cowplot_1.2.0          
## [124] shinyjs_2.1.1