Pharma & Healthcare

Compute relative binding-free-energy differences between two ligands. The gold-standard quantitative step in lead optimisation — ranks closely related candidates without re-running docking.

Estimate the binding energy of a lead compound against an unintended secondary protein target. Predicts side effects before pre-clinical.

Compute electronic excited-state energies for UV-photodynamic therapy, photosensitiser design, and photoswitchable drug candidates. Multi-reference ansatz captures the strong configuration mixing typical of excited states.

Estimate binding energy between a small-molecule ligand and a protein active site. Drug discovery primary screening — replaces or augments molecular docking.

Compute transition-state energies along a candidate biosynthesis or drug-metabolism pathway. Process chemistry and CMC workload.

Predict the lowest-energy conformation of an antibody complementarity-determining region (CDR) loop. Critical for therapeutic-antibody design.

Select an optimal patient cohort that satisfies trial eligibility constraints (inclusion / exclusion criteria, demographic balance).

Select an informative subset of SNP markers from a genome-wide association study under linkage-disequilibrium constraints. Personalised-medicine and population-genetics workload.

Assign surgical cases to operating rooms and staff shifts under equipment, anaesthesiologist, and turnover constraints. Hospital-operations workload.

Predict the lowest-energy 3-D fold of a short peptide on a lattice. Used for de novo peptide design, antibody loop modelling, and as a pre-screen before all-atom MD.

Smith-Waterman style local alignment between two short biological sequences encoded as a lattice QUBO. Genome assembly and variant calling primitive.

Place rotamers for each residue side chain to minimise van der Waals clashes given a fixed backbone. Inner loop of homology modelling and structure refinement.

Predict Absorption, Distribution, Metabolism, Excretion, Toxicity from computed molecular descriptors. Filters compound libraries before any wet-lab assay runs.

Predict adverse interactions between co-administered drugs by computing a quantum kernel over paired molecular fingerprints. Required for polypharmacy clinical decision support.

Quantum-kernel similarity between molecular descriptors for virtual screening of large compound libraries. Cheminformatics workload.

Classify single-lead ECG segments (normal / atrial fibrillation / ventricular tachycardia, …) after wavelet feature extraction. Wearable / Holter monitor pipeline.

Compress medical images to a quantum latent representation, then flag anomalies (e.g. rare lesions, tumour edges) via reconstruction loss. Complements the supervised classifier when labelled data is scarce.

Classify medical images (X-ray, MRI, CT) after classical dimensionality reduction. Diagnostic-AI auxiliary workload.