Skip to contents

Leduc et al, 2025 (bioRxiv): Principles of protein abundance regulation across single cells in a mammalian tissue

leduc2025.Rd

Single-cell samples were prepared using the nPOP glass-slide sample preparation method. Proteomics data were acquired using the timsControl 3.1 on a timsTOF SCP mass spectrometer. The dataset contains quantitative information on 4,184 single cell from murine trachea. The data are available at the peptide, and protein levels. The paper investigates the protein regulation by translation and protein clearance using a matching single cell RNA Sequencing data.

Usage

leduc2025

Format

A QFeatures object with 11 assays, each assay being a SingleCellExperiment object:

  • Assay 1-4: unprocessed peptide data containing quantitative data coming from four replicates.

  • Assay 5-8: processed peptide data containing quantitative data coming from four replicates.

  • proteins_absolute: protein data containing quantitative data for 2,339 proteins and 4,184 single-cells.

  • proteins_relative: protein data containing quantitative data for 2,331 proteins and 4,184 single-cells. Relative abundance values against the protein mean across all samples.

  • proteins_imputed: protein data containing quantitative data for 667 proteins and 4,184 single-cells with k-nearest neighbors (KNN) imputation and comBat batch-correction.

The colData(leduc2025()) contains cell type and batch annotations that are common to all assays.

Source

The data were downloaded from the Slavov Lab website via a shared Zenodo folder folder. The raw data and the quantification data can also be found in the MassIVE repository MSV000098940: ftp://massive-ftp.ucsd.edu/v10/MSV000098940/.

The data used and generated as part of the scpdata repackaging is available in the Zenodo archive.

Acquisition protocol

The data were acquired using the following setup. More information can be found in the source article (see References).

  • Cell isolation: CellenONE cell sorting.

  • Sample preparation performed using the nPOP sample preperation method. cell lysis (DMSO) + trypsin digestion + mTRAQ reagents.

  • Separation: online nLC (Vanquish Neo UHPLC with a 25cm x 75um IonOpticks Aurora Series UHPLC column (AUR2-25075C18A).

  • Mass spectrometry: timsTOF SCP mass spectrometer (MS1 scan range = 100-1700 m/z; MS2 scan range = 300-1000 m/z).

  • Data analysis: DIA-NN(v1.9.0).

Data collection

The Peptide data were collected from a Zenodo folder that is accessible from the SlavovLab website (see Source section).

The folder ('02_raw_reptide_X_singleCell/') contains the following files of interest:

  • r1_peptide.csv: peptide level unprocessed quantitative data for replicate 1.

  • r2_peptide.csv: peptide level unprocessed quantitative data for replicate 2.

  • r3_peptide.csv: peptide level unprocessed quantitative data for replicate 3.

  • r4_peptide.csv: peptide level unprocessed quantitative data for replicate 4.

('03_QuantQC_objects/') contains the following files of interest:

  • r1_5day_male.RData: peptide level processed quantitative data for replicate 1 and row data information for peptide to protein mapping.

  • r2_5day_female.RData: peptide level processed quantitative data for replicate 2 and row data information for peptide to protein mapping.

  • r3_10day_male.RData: peptide level processed quantitative data for replicate 3 and row data information for peptide to protein mapping.

  • r4_10day_female.RData: peptide level processed quantitative data for replicate 4 and row data information for peptide to protein mapping.

('04_Gene_X_SingleCell_and_annotations/') contains the following files of interest: sc_protein_absolute.csv: protein level quantitative data with absolute protein concentrations. sc_protein_relative.csv: protein level quantitative data with relative protein abundance. sc_protein_annotations.csv: sample and batch annotations.

We also obtained imputed and batch corrected protein level data. To do that we run the script get_protein_Data.R based on the original script 02_cell_type_assign.R in the github repository of the article.

We formatted the peptide level quantification tables so that columns match with those of the annotation and protein level data. Each peptide level data for replicates turned into SingleCellExperiment objects including row data information that is extracted from shared QuantQC objects. All the peptide level assays used to generate QFeatures object.

Protein level data were taken from '04_Gene_X_SingleCell_and_annotations/' folder. Imputed and batch effect corrected protein level data is generated using get_protein_Data.R script. The protein assays were formatted as SingleCellExperiment objects and the sample metadata were matched to the column names. The objects are then added to the QFeatures object and the rows of the peptide data are linked to the rows of the corresponding protein data based on the protein sequence information through an AssayLink object.

References

Leduc A, et al., Principles of protein abundance regulation across single cells in a mammalian tissue, bioRxiv, doi: 10.1101/2025.09.17.676955 (link to article).

Examples

# \donttest{
leduc2025()
#> see ?scpdata and browseVignettes('scpdata') for documentation
#> loading from cache
#> An instance of class QFeatures (type: bulk) with 11 sets:
#> 
#>  [1] pep_raw_prep1: SummarizedExperiment with 5635 rows and 1147 columns 
#>  [2] pep_raw_prep2: SummarizedExperiment with 6820 rows and 1002 columns 
#>  [3] pep_raw_prep3: SummarizedExperiment with 5624 rows and 1139 columns 
#>  ...
#>  [9] proteins_relative: SummarizedExperiment with 2331 rows and 4184 columns 
#>  [10] proteins_absolute: SummarizedExperiment with 2339 rows and 4184 columns 
#>  [11] proteins_imputed: SummarizedExperiment with 667 rows and 4184 columns 
# }