iDeepLC: A deep Learning-based retention time predictor for unseen modified peptides with a novel encoding system

Overview

iDeepLC is a deep learning-based tool for retention time prediction in proteomics.

Features

• Retention Time Prediction: Predict retention times for peptides, including modified ones.
• Fine-Tuning: Fine-tune the pre-trained model for specific datasets.
• Visualization: Generate scatter plots and other figures for analysis.

installation

Intall the package using pip:

pip install iDeepLC

Usage

Graphical user interface (GUI)

If you prefer not to install Python or any dependencies, you can use the standalone iDeepLC GUI for Windows.
This is a single .exe file that runs without any installation.

How it works

• When you run the .exe, a terminal window will first appear.
  This terminal acts as the logger for the GUI, showing progress and messages as the program runs.
• Any results and generated figures will be saved in the same folder where the .exe file is located.

Running the executable

1. Download the .exe file from the latest release.
2. Double-click the file to run it.
3. If Windows shows a security message:
   • "Windows protected your PC" — this is a standard warning for applications not signed with a commercial certificate.
   • Click More info and then Run anyway to start iDeepLC.

     This warning appears because the executable is built by the developers without a paid code-signing certificate.
     The file is safe if downloaded from the official GitHub release page.

CLI

The iDeepLC package provides a CLI for easy usage. Below are some examples:

Prediction

ideeplc --input <path/to/peptide_file.csv> --save

Fine-tuning

ideeplc --input <path/to/peptide_file.csv> --save --finetune

Calibration

ideeplc --input <path/to/peptide_file.csv> --save --calibrate

Example

ideeplc --input ./data/example_input/Hela_deeprt --save --finetune --calibrate

For more detailed CLI usage, you can run:

ideeplc --help

Input file format

The input file should be a CSV file with the following columns:

• seq: The amino acid sequence of the peptide. (e.g., ACDEFGHIKLMNPQRSTVWY)
• modifications: A string representing modifications in the sequence. (e.g., 11|Oxidation|16|Phospho)
• tr: The retention time of the peptide in seconds. (e.g., 1285.63)

For example:

NQDLISENK,,2705.724
LGSPPPHK,3|Phospho,2029.974
RMQSLQLDCVAVPSSR,2|Oxidation|4|Phospho,4499.832

Citation

If you use iDeepLC in your research, please cite our paper:

📄 iDeepLC: A deep Learning-based retention time predictor for unseen modified peptides with a novel encoding system
🖊 Alireza Nameni, Arthur Declercq, Ralf Gabriels, Robbe Devreese, Lennart Martens, Sven Degroeve , and Robbin Bouwmeester
📅 2025
🔗 DOI