remove_proteins_by_name: Completely remove proteins, and all their peptides, that...

In ftwkoopmans/msdap: Mass Spectrometry Downstream Analysis Pipeline

Completely remove proteins, and all their peptides, that match some filter from the dataset

Description

Completely remove proteins, and all their peptides, that match some filter from the dataset

Usage

remove_proteins_by_name(
  dataset,
  irt_peptides = FALSE,
  fasta_contaminants = FALSE,
  regular_expression = "",
  gene_symbols = NULL,
  print_nchar_limit = 150
)

Arguments

dataset	the dataset to filter. Note that prior to calling this function, you must have applied import_fasta() such that this function has access to the fasta headers of each proteingroup
irt_peptides	try to find the irt spike-in peptides in the fasta file header. This requires inclusion of the IRT peptides in the samples, using the IRT FASTA during Spectronaut/DIA-NN/x data search, including the IRT FASTA in import_fasta(). This specifically matches all proteins where the fasta header contains any of; "|IRT|", "IRT_KIT", "Biognosys iRT" (case insensitive). default:FALSE
fasta_contaminants	remove proteins that are flagged as a contaminant in the fasta files. Note that this only protein matches from specific "contaminants" FASTA files that were included in your DIA-NN/MaxQuant/etc. search. This specifically matches all proteins where the protein identifier contains any of; "con_", "_con", "|crap-" (case insensitive). default:FALSE
regular_expression	careful here, regular expressions are powerful but complex matching patterns. Here you can provide a 'regex' that is matched against the fasta header(s) of a proteingroup. case insensitive!
gene_symbols	an array of gene symbols that are to be matched against the fasta header(s) of a proteingroup. Symbols must be at least 2 characters long and match exactly, but matching is case insensitive
print_nchar_limit	max number of characters for the fasta headers (of removed proteins) that are shown in the log

Examples

## Not run: 
### example 1:
# If you included a contaminant FASTA in DIA-NN/MaxQuant/etc.,
# you can use this function remove these proteins from the dataset before
# running the MS-DAP analysis_quickstart() function.
#
# First, use DIA-NN to analyze raw files while providing as FASTA files
# 1) The uniprot fasta file(s) that describe your experiment's proteome
#   (e.g. uniprot Human proteome, both the canonical and additional files)
# 2) Check the "Contaminants" box in DIA-NN to include the cRAP proteins
#
# Next, we can use MS-DAP to import this dataset and remove the contaminant proteins.

# I) import the dataset as per usual
library(msdap)
dataset = import_dataset_diann(filename = "C:/data/report.parquet")

# II) import all fasta FASTA files that were used in DIA-NN
# Importantly, you have to include all FASTA files in a single import_fasta() call.
# Note that this includes the contaminant FASTA that is bundled with DIA-NN,
# but only if it was used during DIA-NN analysis.
dataset = import_fasta(dataset, files = c(
  "C:/uniprot/2024_01/UP000005640_9606.fasta",
  "C:/uniprot/2024_01/UP000005640_9606_additional.fasta",
  "C:/DIA-NN/1.9.1/camprotR_240512_cRAP_20190401_full_tags.fasta"
))

# III) If so desired, remove contaminant proteins
# If you want to remove all the cRAP proteins up-front, you can completely remove
# them from the dataset using a regular expression matched against FASTA headers.
# Proteins removed by this function will be fully erased from the dataset,
# i.e. matches that are printed to the console will not be used in any downstream step.
dataset = remove_proteins_by_name(dataset, fasta_contaminants = TRUE)

# note: if the fasta_contaminants option does not catch all proteins
# that you intend to remove, for example when you are using a different contaminant
# FASTA, you can add additional filters using the "regular_expression" parameter.


### example 2: remove keratins and IGGs
# This example uses a regular expression, matched against uniprot fasta headers
# (particularly useful for IP experiments);
dataset = remove_proteins_by_name(
  dataset,
  regular_expression = "ig \\S+ chain|keratin|GN=(krt|try|igk|igg|igkv|ighv|ighg)"
)

## End(Not run)

ftwkoopmans/msdap documentation built on March 5, 2025, 12:15 a.m.