We have developed the KvantifyQDK (Quantum Development Kit), which is designed to empower computational chemists and drug-discovery scientists. The KvantifyQDK provides a streamlined and intuitive interface to enable direct integration with classical-quantum chemistry methodologies through embedding techniques, facilitating realistic, hybrid quantum-classical workflows. It utilizes a generic interface for all quantum backends, making our customers’ applications fully platform-agnostic. Applications set up with KvantifyQDK can be run on any supported physical quantum computer or any of the bundled simulators, and it is trivial to switch between them. We are currently gathering feedback on the product, prior to its launch.
As part of our quantum efforts, we have developed the proprietary algorithm FAST-VQE. This is a novel and exceptionally adept algorithm for Noisy Intermediate-Scale Quantum (NISQ) devices, which delivers high-performing and cost-effective calculations of small molecules. We have included FAST-VQE in our framework, thus bringing the usage of quantum computing into the realm of realistic, industry-relevant applications.
Our classical computing efforts have focused on developing modules for extracting reaction-rate binding constants, and this has led to the development of our first product, Koffee Unbinding Kinetics. This innovative product ranks ligands based on residence time and allows users to calculate unbinding kinetics in minutes, thus complementing and reducing costly and time-consuming laboratory experiments. Unbinding kinetics is a critical parameter in the screening and selection of new drug candidates. The product adds value already today when used in a classical-computing framework, and our algorithm is designed to demonstrate quantum advantage as hardware capabilities evolve and mature toward the fault-tolerant era. Product trials are currently ongoing, and customer feedback is crucial for further refinement of the product’s applicability.
To guide potency optimization in drug discovery it is important to be able to measure how strongly a drug binds to its target. Existing, traditional methods either lack speed or precision, and in response to the urgent call for new methods in this space, we developed an algorithm for ranking ligands based on binding affinity. We launched this as our second product, Koffee Binding Affinity. Based on several customer trials, we have verified that our method performs on par with competitive solutions, and at the same time it is significantly faster and more user-friendly. We are currently refining the tool based on customer feedback and market insights.
