កង់ទិច · អត្ថបទ
Quantum Computers Predict Protein Water Sites for Drug Design on IBM Chips
Quantum Computing Report ·

មើលត្រួសៗ
Researchers from Qubit Pharmaceuticals, Sorbonne Université, and Q-CTRL used IBM quantum processors to predict where water molecules sit inside drug-target protein pockets. The study appeared in a peer-reviewed journal, and the team ran hardware tests with up to 123 qubits.
ចំណុចសំខាន់ៗ
- The team turned water-placement modeling into an optimization problem that quantum computers can attempt.
- On 11 test cases using IBM Heron chips, the quantum method found the verified best water arrangements every time.
- For the largest 116-qubit case, the quantum method found the best answer about 9% of the time, versus 2% for a classical method.
- A 123-qubit case on a newer IBM chip beat the exact classical solver CPLEX, which hit a three-hour timeout.
ហេតុអ្វីវាសំខាន់
Predicting where water sits in drug-target pockets helps scientists design medicines, and this shows quantum computers may one day help with that work.
ប្រភពដើម
Qubit Pharmaceuticals, Sorbonne Université, and Q-CTRL Demonstrate 3D Protein Hydration-Site Prediction on IBM Heron Processors
Quantum Computing Report · Nov 21, 17:00 KST
Read on Quantum Computing Report ↗AI summarized the original article. Check the original for accuracy. Copyright belongs to Quantum Computing Report.
បន្ថែមពីវិស័យនេះ
- Sorbonne and Qubit Pharmaceuticals Unveil VULCAN to Cut Quantum Chemistry Measurement Costs
Researchers from Sorbonne Université and the company Qubit Pharmaceuticals introduced VULCAN, a hybrid quantum-classical method for calculating molecular electronic structure. It aims to reach chemical accuracy while needing far fewer quantum measurements than the current ADAPT-VQE approach. The details are in an arXiv preprint.
Quantum Computing Report · Nov 21, 17:14 KST - Qubit Pharmaceuticals and Sorbonne Université Speed Up Quantum Chemistry Operator Pool Checks
Researchers from Qubit Pharmaceuticals and Sorbonne Université published a peer-reviewed paper in Communications Physics on a faster way to build and check the operator sets used in quantum chemistry calculations. The method checks completeness with a matrix rank test instead of costly nested calculations. The authors tested it with GPU simulations of up to 26 qubits and released their verification code as open source.
Quantum Computing Report · Nov 21, 17:10 KST - QuEra and Harvard Researchers Unveil Faster Way to Benchmark Quantum Circuits
Researchers from QuEra, Harvard, the University of Maryland, UC Berkeley, and ETH Zürich built a faster computer-based way to check how accurate certain future quantum circuits are. Their method avoids the huge computing cost of simulating every step. They tested it on circuits with up to 256 qubits (quantum bits, the basic units of quantum computers).
Quantum Computing Report · Nov 21, 16:36 KST - Harvard, QuEra, and Caltech Propose Quantum Magic State Factory With 10-40x Lower Cost
Researchers from Harvard, QuEra Computing, Caltech, the University of Maryland, and MIT proposed a new design for building magic states, the special quantum states needed for fault-tolerant quantum computers. They say it cuts space and time costs by 10 to 40 times compared with current methods. The design uses a new type of error-correcting code and was mapped onto neutral-atom hardware.
Quantum Computing Report · Nov 21, 16:31 KST