EasyPQP: Simple library generation for OpenSWATH

EasyPQP is a Python package that provides simplified and fast peptide query parameter generation for OpenSWATH. It can process input from MSFragger, Sage or other database search engines in pepXML/idXML/tsv format. Statistical validation can be conducted either using PyProphet or PeptideProphet/iProphet. Retention times and ion mobilities are calibrated using an internal or external standard. In addition to a cumulative library, run-specific libraries are generated for non-linear RT alignment in OpenSWATH. For generation of PTM specific libraries that utilizes a unimod.xml database, you can further restrict the unimod.xml database file for modifications and site-specificities of interest. It also supports in-silico library generation.

Installation

We strongly advice to install EasyPQP in a Python virtualenv. EasyPQP is compatible with Python 3.

Install the development version of easypqp from GitHub:

    $ pip install git+https://github.com/grosenberger/easypqp.git@master

Full Installation

To install all optional features:

    $ pip install easypqp[all]

This will install the easypqp_rs package, which provides the in-silico library generation feature and pyprophet for statistical validation.

Running EasyPQP

EasyPQP is not only a Python package, but also a command line tool:

   $ easypqp --help

or:

   $ easypqp convert --help
   $ easypqp convertpsm --help
   $ easypqp convertsage --help
   $ easypqp library --help
   $ easypqp insilico-library --help
   $ easypqp reduce --help
   $ easypqp filter-unimod --help
   $ easypqp openswath-assay-generator --help
   $ easypqp openswath-decoy-generator --help
   $ easypqp targeted-file-converter --help

Generating an In-Silico Library

The in-silico library generation feature is included if you installed EasyPQP with the [all] or [rust] extras (to install the easypqp_rs package).

To generate an in-silico library, you can use the insilico-library command. For example:

   $ easypqp insilico-library --fasta your_proteome.fasta --output_file insilico_library.tsv

For more information on the parameters and JSON configuration file, see the Configuration Reference

[!NOTE] If no retention_time, ion_mobility, or ms2_intensity fields are provided under dl_feature_generators in the config, pretrained models will be automatically downloaded and used. The current default pretrained models used are:

• RT: rt_cnn_tf - A CNN-Transformer model trained on the ProteomicsML repository RT dataset. This model is based on AlphaPeptDeep’s CNN-LSTM implementation, with the biLSTM replaced by a Transformer encoder.
• CCS: ccs_cnn_tf - A CNN-Transformer model trained on the ProteomicsML repository CCS dataset. This model is also based on AlphaPeptDeep’s CNN-LSTM implementation, with the biLSTM replaced by a Transformer encoder.
• MS2: ms2_bert - A BERT-based model retreived from AlphaPeptDeep’s pretrained models.

If you want just a standalone portable rust binary, you can download one from the easypqp-rs releases page.

Docker

EasyPQP is also available from Docker (automated builds):

Pull the development version of easypqp from DockerHub (synced with GitHub):

    $ docker pull grosenberger/easypqp:latest