Quantum optimization shows limits in drug-response modeling on trapped-ion hardware
Feasibility and optimum recovery in warm-start quantum optimization for a drug-response model on a trapped-ion processor
Emerging Technologies
Summary
This paper looks at using a quantum optimization method to predict how drugs affect a simple biological system. The authors found that the quantum method sometimes finds valid solutions but often does not improve over random guesses or classical techniques. They tested their approach on real trapped-ion quantum computers and simulations and found that classical greedy and annealing methods solved the problems more reliably. The research shows the challenges of current quantum hardware and algorithms in optimizing drug-response models.
What this means in practice
- •For quantum computing developers: Evaluate quantum hardware performance and algorithm choice for combinatorial optimization related to biological data modeling.
- •For pharmaceutical data teams: Use benchmarking insights to understand current limits of quantum methods when modeling drug interactions at small scales.
Authors
Tanzir Hossain, Rajib Rana, Prabal Datta Barua, Abu Ali Ibn Sina, Niall Higgins, Pascal Elahi, Robert Sang, Bjorn W. Schuller
Abstract
On a seven-compound drug-response model, warm-start quantum approximate optimization (QAOA) on IonQ Forte-1 returned valid assignments more often than random bitstrings, but this alone did not show effective optimization. Ideal QAOA raised optimum probability above uniform feasible sampling in only four of twelve reference circuits. Hardware often fell below its own noiseless circuits, while greedy search solved all hardware models within 200 objective evaluations. Expanded simulations showed a gain over feasible sampling in 28 of 35 CAMA-1 panels and none of four 647-V panels. Annealing solved all these panels in every seed. A Grover mixer preserved feasibility and improved optimum probability over feasible sampling in all ten tested models. We analyzed 25 completed tasks containing 5,300 shots from 18 circuits and 14 instances. The encodings use 6-35 qubits and at most 4,900 feasible assignments, which we enumerated to establish exact optima. Noiseless references now cover both the original six circuits and six wider circuits. At 35 qubits, with 4,900 feasible assignments, ideal feasibility was 35.85%, compared with 7 of 200 valid hardware outputs. Its ideal optimum probability was below both sampling controls. The circuits sample assignments in a model built from measured single-agent and pairwise responses.