pnnl.diann.utils

The goal of pnnl.diann.utils is

1. A set of functions to analyze DIA-NN outputs and convert them to MSnSet objects for downstream analysis with MSnSet.utils.
2. Assembling paths for analyzing files directly from DMS.
3. (in-progress) General workflow for plexDIA analysis.

Overview of `pnnl.diann.utils` package workflow. Package functions are in `monospace`.

Installation

You can install the development version of pnnl.diann.utils from GitHub with:

# install.packages("devtools")
devtools::install_github("PNNL-Comp-Mass-Spec/PNNL.DIA.utils")

Example datasets

Example gene group (gg) and protein group (pg) matrices, and precursors (pq) report are found in buildinst/extdata/.

• diann_report.tsv is example data from diann_rpackage
• QC_Mam matrices are from \\proto-8\Lumos02\2023_4\QC_Mam_19_01_b_DIA_trap_60min_19Dec23_Pippin_WBEH-23-05-18\DNN202312200344_Auto2255735

Converting Protein Groups Matrix to MSnSet

library(pnnl.diann.utils)


pg_matrix_path <- system.file("extdata",
                              "QC_Mam_19_01_b_DIA_report.pg_matrix.tsv",
                              package = "pnnl.diann.utils")


pnnl.diann.utils::pgmatrix_to_msnset(pg_matrix_path)
#> MSnSet (storageMode: lockedEnvironment)
#> assayData: 5550 features, 1 samples 
#>   element names: exprs 
#> protocolData: none
#> phenoData
#>   sampleNames:
#>     QC_Mam_19_01_b_DIA_trap_60min_19Dec23_Pippin_WBEH-23-05-18
#>   varLabels: dataset
#>   varMetadata: labelDescription
#> featureData
#>   featureNames: A0A075B5S1;A0A075B5S9;A0A0B4J1J4;A0A0B4J1J5
#>     A0A087WNP0;A0A3B2W7J0;E9PX25;E9PZY4;J3QP73;K9J7E2;L7N2C4 ... S4R1W5
#>     (5550 total)
#>   fvarLabels: Protein.Ids Protein.Names Genes First.Protein.Description
#>   fvarMetadata: labelDescription
#> experimentData: use 'experimentData(object)'
#> Annotation:  
#> - - - Processing information - - -
#>  MSnbase version: 2.26.0

Assembling file paths from Data Package number

While running DIA-NN from kaiju or other PNNL servers, DIA-NN can pull files directly from DMS, saving time and redundancy incurred from copying LC-MS files over with a minimal impact on overall DIA-NN analysis time. To this end an absolute file path needs to be provided to DIA-NN for each LC-MS file. dms_dataset_path assembles the paths for all files in a DMS Data Package. The function requires PNNL.DMS.utils and access to PNNL network.

paths <-pnnl.diann.utils::dms_dataset_path(
  data_package_num = 5293,
  file_type = ".raw",
  diann_console = TRUE
  )


paths[1:3,]
#> [1] "--f \"\\\\proto-9\\Eclipse02\\2023_4\\KidsFirst_T-ALL_PATEAK_Prot_061_FAIMS_Merry_DI_16Oct23_WBEH-23-07-14\\KidsFirst_T-ALL_PATEAK_Prot_061_FAIMS_Merry_DI_16Oct23_WBEH-23-07-14.raw\""
#> [2] "--f \"\\\\proto-9\\Eclipse02\\2023_4\\KidsFirst_T-ALL_PATGMS_Prot_063_FAIMS_Merry_DI_16Oct23_WBEH-23-07-14\\KidsFirst_T-ALL_PATGMS_Prot_063_FAIMS_Merry_DI_16Oct23_WBEH-23-07-14.raw\""
#> [3] "--f \"\\\\proto-9\\Eclipse02\\2023_4\\KidsFirst_T-ALL_PARLST_Prot_002_FAIMS_Merry_DI_16Oct23_WBEH-23-07-14\\KidsFirst_T-ALL_PARLST_Prot_002_FAIMS_Merry_DI_16Oct23_WBEH-23-07-14.raw\""

Note the outputs above are already formatted for the DIA-NN Console Command. They can then be saved (see below) and pasted into the Console Commands box on DIA-NN. Setting diann_console = FALSE will return the absolute LC-MS file paths only.

write_tsv(x = diann_raw_paths,
          escape = "none",
          col_names = FALSE, 
          file = "DATA/PATH.txt")