Open Discovery Challenge: How to Build a Verifiable Judge for AI-Designed Malaria Drugs
Open Discovery Challenge: How to Build a Verifiable Judge for AI-Designed Malaria Drugs Generative models can propose thousands of plausible molecules in a day. The harder question is no longer whether an AI can draw a molecule. It is whether anyone can tell if that molecule is potent, selective, safe enough to investigate, and possible to synthesize. That is the premise of the Open Discovery…
The Open Discovery Challenge aims to create a verifiable judge for AI-designed malaria drugs. The challenge focuses on a malaria parasite enzyme called PfDHODH. A successful drug must inhibit the parasite enzyme without affecting the human counterpart, survive in the environment, and cross cell membranes. The challenge provides a scoring system with six axes, including whole-cell activity, binding, selectivity, ADMET profile, novelty, and synthesis.
Validation of the scoring system revealed several issues, such as toxicity gate errors, size bias in molecular weight, and silent errors in novelty scoring. These defects highlight the difficulty of building a fair scientific judge for AI-generated molecules.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
