Quick Start

RIANA enables protein turnover data analysis from mass spectrometry-based proteomics experiments. Supports multiple labeling strategies, including labeled amino acid and heavy water.
Author

Edward Lau

Published

April 25, 2023

Abstract
This page provides the documentation for the RIANA package, which is a tool for the analysis of stable isotope labeling experiments for protein turnover measurements.

Installing RIANA

Install Python 3.7+ and pip. See instructions on Python website for specific instructions for your operating system.

Riana can be installed from PyPI via pip or directly from GitHub. We recommend using a virtual environment.

$ pip install riana

Launch riana as a module (Usage/Help):

$ python -m riana

Alternatively as a console entry point:

$ riana

To test that the installation can load test data files in tests/data:

$ pip install tox
$ tox

To run the Riana test dataset (a single fraction bovine serum albumin file from a Q-Exactive) and write the results to a specified out/test/ directory

$ python -m riana integrate tests/data/sample1/ tests/data/sample1/percolator.target.psms.txt -q 0.1 -i 0,1,2,3,4,5 -o out/test/

Processing Datasets

Via Snakemake

RIANA comes with a Snakemake pipeline which is the recommended way to run RIANA. To configure, edit the provided config_template.yaml to specify the location of Comet and Percolator executables. Then run Snakemake with the provided Snakefile, e.g.,:

$ snakemake -c -s ./Snakefile -d out/snakemake_test --configfile ./config_template.yaml

Refer to Snakemake documentations for details on configuring snake files.

Curve plotting

Note that the provided Snakefile settings do not automatically output plotted curves or pre-integration intensity files by default. To output these files, edit the Snakefile to add -w and p to the shell commands under the integrate and fit rules, respectively.

Below is the config_template.yaml:

# Config

# Data location. Add as many lines as needed, sample names should be named after
# labeling time points (e.g., time12 for 12 days/12 hours of experiment)
data:
  time0: /path/to/time0/folder
  time1: /path/to/time1/folder
  time3: /path/to/time3/folder
  time6: /path/to/time6/folder

# Paths to the comet executable, the comet params file, the database, and the percolator executable
# The Snakefile assumes riana can be executed in the shell by riana
paths:
  comet: /path/to/comet_source_2020013/comet.exe
  comet_params: /path/to/params/comet.params
  fasta: /path/tp/database.fas
  percolator: /path/to/percolator

# Integration and fitting parameters
params:
  model: simple     # fitting model (simple, guan, fornasiero)
  label_type: hw        # labeling type (aa or hw)
  isotopomers: 0,1,2,3,4,5   # isotopomer to integrate (0,1,2,3,4,5 for deuterium; 0,6 for heavy aa)
  mass_tol: 15         # mass tolerance in ppm for integration (e.g., 25)
  ria_max: 1          # final precursor ria
  kp: 10              # guan model parameter (kp) or fornasiero model parameter (b)
  kr: 0.15              # fornasiero model parameter (a)
  rp: 5.52              # fornasiero model parameter (r)
  depth: 1             # minimum number of data points
  aa: K                 # which amino acid carries heavy label; only relevant if label_type is aa
  mass_defect: D        # which isotope mass defect to use (D, C13, SILAC)

# Number of threads
threads:
  comet: 8
  riana: 4
  fitcurve: 12

Manual Runs

Alternatively, RIANA can be directly called from command line given the paths to the folder of the mzML files and the Search/Percolator results. To run the turnover analysis pipeline manually, prepare the following expected input files:

  • RIANA was tested on the percolator output file from Crux Tide/Percolator or standalone Comet/Percolator.

  • The following workflow has been tested for both amino acid and heavy water labeling data gathered on a Q-Exactive instrument:

  • Convert raw files to mzML, using pwiz 3.0 msconvert in command line, with the following option:

    • --filter "peakPicking vendor"

  • Download the latest Crux distribution

  • Run Tide index with the following options:

    • --digestion partial-digest
    • --missed-cleavages

  • Run Tide search for each experimental time point separately with the following options:

    • --isotope-error 1,2 (for HW) or 6,12 (for AA)
    • --compute-sp T
    • --mz-bin-width 0.02
    • --mz-bin-offset 0.0
    • --precursor-window 20
    • --precursor-window-type ppm

  • Run Percolator with the following options:

    • --protein T
    • --decoy-prefix DECOY_

  • Run RIANA integrate for each experimental time point separately.

  • Run RIANA fit

Specifying paths

Note that the mzml_path argument should point to the directory where all mzML fractions (fraction1.mzML, fraction2.mzML, etc.) for each labeling time point are located, whereas the id_path argument should point to a percolator.target.psms.txt file from the output of searching and post-processing all the mzML files in that labeling time point together

Output

RIANA integrate outputs a {sample}_riana.txt file for each sample, which contains the Percolator output appended with additional columns, e.g., m0, m6, m12, which contains the integrated areas of each isotopomer of each peptide over a retention time range in the mass spectrometry data. The output columns depend on the isotopomer indices specified.

In addition, the {sample}_riana_intensities.txt files contain the individual intensity data of each isotopomer in each retention time point prior to integration.

RIANA fit outputs a riana_fit_peptides.csv file which contains the best-fit k, R2, and dk for each peptide-charge combination.

If the flag --plotcurves is set, RIANA additional outputs a graphical representation of each fitted curve:

Best fit kinetic curve for a particular peptide across multiple labeling time points.