simsi-transfer 0.2.0

Software-assisted reduction of missing values in phosphoproteomics and proteomics isobaric labeling data using MS2 spectrum clustering

SIMSI-Transfer

Tool for increasing PSM gain from MaxQuant output file. Requires "Experiment" set in MaxQuant!

For testing SIMSI-Transfer after installation, we recommend downloading the TMT11 MS2 raw files from this publication:
Thompson, A., Wölmer, N., Koncarevic, S., Selzer, S. et al., TMTpro: Design, Synthesis, and Initial Evaluation of a Proline-Based Isobaric 16-Plex Tandem Mass Tag Reagent Set. Analytical Chemistry 2019, 91, 15941–15950. doi:10.1021/acs.analchem.9b04474

The MaxQuant and SIMSI-Transfer results of the processed dataset can be downloaded from the following DOIs:

• MaxQuant output (required as input for SIMSI-Transfer): 10.5281/zenodo.6365902
• SIMSI-Transfer output: 10.5281/zenodo.6365638

Running SIMSI-Transfer using the GUI

On Windows, you can download the SIMSI-Transfer_GUI_windows.zip from the latest release, unzip it and open SIMSI-Transfer.exe to start the GUI (no installation necessary).

Alternatively, on all platforms, first install SIMSI-Transfer as explained below. Then install PyQt5 (pip install PyQt5) and run:

python gui.py

Running SIMSI-Transfer from the command line

First install SIMSI-Transfer as explained below, then run SIMSI-Tranfer:

python -m simsi_transfer --mq_txt_folder </path/to/txt/folder> --raw_folder </path/to/raw/folder> --output_folder </path/to/output/folder>

Installation

SIMSI-Tranfer is available on PyPI and can be installed with pip:

pip install simsi-transfer

Alternatively, you can install directly from this repository:

git clone https://github.com/kusterlab/SIMSI-Transfer.git
pip install .

Building the GUI on Windows

Create a conda environment

Try importing conda_environment.yml in the Anaconda environment tab.

If that does not work, try the following:

1. Set up a new environment, either through the Anaconda UI, or by running the following on the command line:

conda create -n simsi_transfer_gui python=3.8
activate simsi_transfer_gui

2. There are some caveats with installing the dependencies. We want to avoid dependence on the MKL (Math Kernel Library) package by numpy/scipy, as this blows up the size of the .exe file over 200MB (see here).

conda install -c conda-forge nomkl numpy pandas pyqt pyinstaller
conda install -c bioconda pyteomics

Building a self-contained executable

Use the build_gui.bat script to create a self-contained executable.

Reducing size of the executable

Download UPX (https://upx.github.io/) to reduce the DLL file sizes, change the path in build_gui.bat to point to the UPX folder.