Skip to content

OpenADMET/ExpansionRx-Challenge-Tutorial

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

17 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ExpansionRx-Challenge-Tutorial

Logo

This repo provides a guide and example workflows to participate in the ExpansionRx-OpenADMET Blind Challenge, a community-driven initiative to benchmark models for predicting ADMET (Absorption, Distribution, Metabolism, Excretion, Toxicity) properties in drug discovery on a unique real world dataset.

In the tutorial notebook you will find an example workflow to train a simple set of models using the provided training data, and generate predictions for submission to the challenge platform on Hugging Face.

We have a dedicated Discord server for Q&A, discussion, and support during the challenge. Our evaluation logic is also open source and available on in this repo. We welcome feedback and community discussion on all aspects the challenge!


🧪 About the Challenge

Participants are tasked with predicting 9 ADMET endpoints for small molecules using real-world drug discovery data provided by ExpansionRx.

Endpoints:

  • LogD
  • Kinetic Solubility (KSol)
  • Mouse Liver Microsomal Clearance (MLM CLint)
  • Human Liver Microsomal Clearance (HLM CLint)
  • Caco-2 Efflux Ratio
  • Caco-2 Permeability (Papp A>B)
  • Mouse Plasma Protein Binding (MPPB)
  • Mouse Brain Protein Binding (MBPB)
  • Mouse Gastrocnemius Muscle Binding (MGMB)

More details on endpoints: Blog Post


📦 Dataset

  • Training Data:
    Available on Hugging Face
    Includes SMILES and ADMET measurements for a series of molecules.

  • Test Data:
    Blinded — predictions must be submitted to the challenge platform. Blinded test data also available on HuggingFace


✅ How to Participate

  1. Follow the tutorial via this repo to learn about the submission process.
  2. Download the public dataset.
  3. Train your model on any or all endpoints.
  4. Submit predictions to the challenge platform (1 per day max, latest counts).
  5. Join the Discord for Q&A and support.

🏆 Evaluation

  • Scored using Mean Absolute Error (MAE) per endpoint.
  • Overall ranking by Macro-Averaged Relative Absolute Error (MA-RAE).
  • Submissions may include external data or pretraining.
  • Submissions can be anonymous if desired
  • Winners announced January 26, 2026.

📅 Key Dates

  • 🗓 Challenge Starts: October 27, 2025
  • Submission Deadline: January 19, 2026
  • 🏁 Winners Announced: January 26, 2026

About

Tutorial for the ExpansionRx-OpenADMET blind challenge

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •