AI2050 H3D in-house workshop

This repository contains all materials and resources from the AI2050 H3D in-house workshop held in June 2025.

Session 1: Ready-to-use AI/ML models for drug discovery with the Ersilia Model Hub

Drug Discovery is a long and expensive process - how can AI/ML methods help?

Why is this particularly relevant in low-resource settings?

What is the Ersilia Model Hub?

Session 2: Predicting ADMET properties for your compounds of interest

ADMET properties are a fundamental consideration in drug discovery pipelines. ADMET refers to Absorption, Distribution, Metabolism, Excretion, and Toxicity. These properties determine the pharmacokinetic and safety profiles of drug candidates and play a critical role in predicting how a compound behaves in the human body.

Examples of ADMET properties are:

  • Blood-brain barrier penetration

  • Cytochrome inhibition

Predicting ADMET properties using the Ersilia Model Hub.

Session 3: Predicting bioactivity for your compounds of interest and selecting candidates to purchase

  • Breakout: mtb? ab? kp?

  • TBD between SA, cytotoxicity and 2D projections

  • Enamine: maybe HLL-100 ? 100k

  • H3D in-house?

Last updated

Was this helpful?