Download PUNCH2 and Embedding Software

We offer two ways to download the PUNCH2 software, allowing you to set it up on your own server or laptop. The software is compatible with Linux, Mac, and Windows operating systems. The prediction functionality is divided into two separate packages for flexibility: the sequence embedding package, called "Embedding," and the predictor package, named "PUNCH2-light." You can download and install these packages locally either via GitHub or Docker. Although both options are available, we recommend using Docker, as it simplifies the setup process by eliminating the need to manually install dependencies or configure the environment.

Regardless of whether you choose GitHub or Docker, you need to follow these steps:

Refer to the README files in the repositories for detailed installation instructions.
Run the Embedding tool to generate sequence embeddings.
Use the generated embeddings as input for the PUNCH2-light software to obtain IDR predictions.

Option 1: Download the Software from GitHub

The source code and installation instructions are available on our GitHub repositories:

Option 2: Use Docker (Recommended)

Alternatively, you can run the software through Docker without manually installing any dependencies. The installation instructions are provided in the GitHub repositories. You can access the Docker images through the following links:

For more detailed information, please refer to the documentation available in the PUNCH2 GitHub repository.

Back to Home

PUNCH

A residue-level Intrinsically Disordered Region/Protein (IDP/IDR) predictor trained on PDB and Disprot Databases.

Home

Download PUNCH & Embedding Software

Intrinsically disordered proteins (IDPs) lack stable 3D structures, making computational prediction challenging. PUNCH web server is a fast, lightweight tool with high accuracy, excelling in benchmarks against top predictors on the CAID3 dataset, offering valuable insights for IDP research.

Citation:
Meng, D., & Pollastri, G. (2025). PUNCH2: Explore the strategy for intrinsically disordered protein predictor. PLOS ONE, 20(3), e0319208. https://doi.org/10.1371/journal.pone.0319208
Meng, D., & Pollastri, G. (2025). Punch: An Interactive Web Server for Predicting Intrinsically Disordered Regions in Protein Sequences. J Mol Biol. 2025;169018. doi:10.1016/j.jmb.2025.169018.