Scientists’ Side Hustle? Using AI and Quantum Computing to Generate New Peptides - WIRED

Researchers at the Technical University of Denmark used a quantum computer from ORCA Computing to enhance the performance of an artificial intelligence model designed to generate novel peptides, according to a report in WIRED. The study aimed to improve both the accuracy and diversity of peptide sequences produced by the AI, which could be used to develop new drugs. The work is presented as a step toward aiding drug discovery for populations that are currently underserved by existing treatments.

The researchers integrated a quantum computing system into the AI's training process, allowing the model to explore a wider range of molecular structures than classical methods typically allow. The enhanced model reportedly produced peptide candidates that were more structurally diverse and had higher predicted synthesizability compared to those generated by classical approaches. The specific improvements in accuracy and diversity were not quantified in the article, but the team stated that the quantum component enabled the model to overcome previous computational bottlenecks.

The path from computationally generated peptide candidates to approved drugs typically takes over a decade and costs billions of dollars. The study does not claim that any of the generated peptides have been tested in biological assays or animal models. The collaboration was between an academic institution and a quantum computing company, reflecting ongoing efforts to find concrete use cases for quantum hardware in scientific domains.