Skip to content
DIMECHAIN

Drug discovery

Quantum Powered Protein DesignMaturity: concept

Machine learning has transformed structure prediction; design — going from a desired function to a sequence — is still much harder, and the gap between a predicted structure and a working protein is routinely surprising. We are looking at conformational search and interaction scoring where classical scoring functions are known to be weak. Nothing here counts without wet-lab validation, which is a field where computational and experimental results diverge often enough that nothing else does.

concept An idea we find credible. Nothing has been built or measured.

What this is

The problem

Designing a protein to fold into a chosen structure and perform a chosen function is an enormous combinatorial search, with a scoring function that is itself approximate.

Where the current approach strains

Machine-learning methods have transformed structure prediction. Design — going from desired function to sequence — remains harder, and the gap between a predicted structure and a working protein is still routinely surprising.

What we are exploring

Conformational search and interaction scoring, in the regions where classical scoring functions are known to be weak.

What would have to be true

Wet-lab validation. Protein design is a field where computational results and experimental outcomes diverge often enough that nothing else counts.

Published as a case study on the legacy site with no experimental validation behind it, which is why it appears here as a concept.

Provenance

Published as a case study on the legacy site. It described an application concept rather than a delivered engagement with measured outcomes, so it lives here with a maturity label instead.

Previously published at https://quzones.com/case-studies/quantum-powered-protein-design/

Where it applies

Related

Want to co-develop this?