Features
Dashboard
Software Development Kit (SDK)
JupyterLab & Kubeflow
Machine Learning & Large Language Models (LLM) Library
Features
Born-Oppenheimer and Car-Parrinello Molecular Dynamics
GPUs and QPUs Acccess per Request
Select Features
Kubernetes Setup with Infrastructure as Code
MQS Containers
Laboratory Connection for ML-driven Design of Experiments and Robotic Lab Operation
Ready to revolutionize your research workflow? The MQS Dashboard offers a visual interface to search through quantum chemistry data of over 200 million molecules. On top of the advanced search capabilities, JupyterLab and Kubeflow are integrated with the whole MQS infrastructure to leverage high performance CPUs, GPUs and QPUs.
Over 200 million molecules Searchable PubChemQC PM6 and QMugs datasets
High performance compute infrastructure JupyterLab and Kubeflow support
The MQS solvation, mapping and measurement containers can be seamlessly integrated with your hybrid classical-quantum calculation pipelines. Create modular and well-defined pipelines with Kubeflow and the MQS containers to develop robust and sustainable prediction pipelines.
The MQS REST-API allows to integrate any tool available in the MQS tool stack into your infrastructure.
Calculate.
Apply high-performance quantum chemistry and molecular dynamics for molecular property prediction to calculate for example solubility and phase equilibria.
Find.
Find promising molecular systems and define a design of experiments for the laboratory.
Develop.
Combine in-silico and experimental data sets to perform modern R&D with Bayesian closed-loop optimisation.
Designed to seamlessly integrate with the MQS Dashboard, our API and SDK allows for numerous possibilities for utilizing and developing advanced analysis workflows. Whether you are a seasoned scientific developer or just getting started, the API and SDK enhances your workflow, providing access to molecular data and high-performance infrastructure. In combination with the integrated JupyterLab and Kubeflow you are able to make use of all tools without any setup overhead.
Partner with MQS to start a collaborative project to apply our software tools and experimental validation for the pharmaceutical, crop science, chemical and material industries.
The workshop is divided up over three flexible dates which are being set with the workshop participants beforehand. Please sign up via workshop@mqs.dk if you are interested in participating. The workshop is made for engineers, researchers and PhD students interested in learning how to program a quantum computer. You will learn the basic operating principles of a quantum computer, and get introduced to the quantum computing architectures and operating systems that are in operation today and in the near future. Through exercises in our course material you will learn quantum computer programming, to perform quantum chemistry calculations and to construct tensor networks in Python using the pennylane, qiskit, pyscf, psi4, cebule and quimb packages.
The use cases we will look into are optimization, machine learning and chemical simulation on a quantum computer. The examples covered will be related to chemical engineering topics but are kept general and can readily be extended to other fields such as logistics or financials. All examples and exercises will be offered on our JupyterHub instance and no installation of any packages are needed on your own devices. You will retrieve a Python onboarding before the workshop starts to be able to work on the workshop exercises without any technical difficulties. The workshop assumes no prior experience with quantum computing, but familiarity with linear algebra and the Python programming language is recommended to be able to follow the presentations and work on the exercises. The workshop is structured as follows:
The workshop will be available online (fully remote) and classes will start at 9.00 AM Central European Time (CET) and last approx. until 2.00 PM CET. Throughout the workshop and additional two weeks you will have access to our online quantum computing for chemistry course book and our self-hosted JupyterHub with exercise Jupyter notebooks, which have been specifically designed to complement the lecture material and to facilitate learning by providing an opportunity to directly apply the theoretical knowledge presented in the workshop. As the online workshop platform we will use the virtual classroom software BigBlueButton which we also host ourself. Again, you will have no hassle to install anything on your computer. Just have pen and paper next to your computer to make notes and experience a cutting-edge workshop teaching the newest developments in quantum computing. The whole platform stack we are using is GDPR compliant.
The instructors will be Kaur Kristjuhan (LinkedIn) and Mark Jones (LinkedIn) from MQS.
Participant outside of Denmark: 1500 EUR excl. tax Participant from Denmark: 11200 DKK excl. tax To sign up for the workshop, please send us an email: workshop@mqs.dk
The MQS team members are highly-interdiscplinary working at the intersection of quantum chemistry, quantum computing, chemical engineering, computer science, digital transformation and laboratory automation.
CEO/CTO
CCO
Quantum Computing Methods Developer and Industrial PhD
Quantum Computing Methods Developer
Full Stack Developer and Computer Scientist
DevOps Engineer and Computer Scientist
Quantum Chemistry, Machine Learning and Quantum Computing Pipelines Developer
Projects Coordinator & Management Support
UI and Web Designer
Lab Automation Engineer
Lab Automation Engineer
Business Developer
Scientific Advisor
Scientific Advisor
Scientific Advisor
Chairman of the Board
Member of the Board
Observer of the Board
Richard P. Feynman is regarded as the person who sparked the idea of a quantum computer with his talk titled "There is Plenty of Room at the Bottom" given at the American Physical Society in Pasadena (December 1959).
A transcript from his talk "There's Plenty Room at the Bottom" conveys Feynman's physical notion of information storage from a volumetric perspective and depicts DNA as an example, where "around 50 atoms are used for one bit of information about the cell". Then Feynman describes how small computers could be built and fabricating electric circuits on the atomistic level although resistance would be a problem. The problem, Feynman mentions, could be solved through the use of superconductivity. At this scale, building a computer with atoms becomes quantum mechanically defined and "We can use, not just circuits, but some system involving the quantized energy levels, or the interactions of quantized spins, etc.". Although Feynman does not use the term "quantum computer" one can already see the vision and follow the line of thought he had.
David Finkelstein acknowledged in his "Space-time Structure in High Energy Interactions" article that it was Feynman who introduced him to the idea that space-time should be quantum mechanically discretized and in this way a "reasonable model for the world is a computer, a giant digital computer" would allow more advanced computations.
In 1979 Paul Benioff constructed a quantum mechanical model of computers where the evolution of a Hamiltonian (energy function) would represent a closed conservative system. Benioff showed in the article "The computer as a physical system: a microscopic quantum mechanical hamiltonian model of computers as represented by Turing machines" that this quantum mechanical model can be calculated by a set of Turing machines which themselves are stationary systems, but collectively, these processing units can simulate dynamic systems.
Richard Feynman's description of how to accurately calculate quantum physical systems on a quantum computer architecture can be seen as the first description of a quantum computer architecture. He submitted the description on the 7th of May 1981 to the International Journal of Theoretical Physics. The article was then published in volume 21, June, 1982. Together with Richard Feynman, David Deutsch is regarded as the father of the idea and detailed description of a quantum computer. David Deutsch described in 1985 his idea of an universal quantum computer. In the proceedings paper "The church–turing principle and the universal quantum computer" he describes a computing machine with a quantum physical architecture and that it can have many remarkable properties which cannot be found in Turing machines. Further, David Deutsch proved also with the Deutsch and Deutsch-Josza algorithm that a quantum computer can give an advantageous speed up in comparison to a classical computing scheme.
Quantum computing has been studied for decades and ranges from information theory to hardware technologies, computational models implemented on specific hardware types which go hand in hand with the mathematical problem formulations being solved on the quantum device.
It has applications in nearly every field that contains or utilizes computations with high complexity. Quantum computers have the potential to impact many aspects of current domains of science, including computer science, mathematics, and chemical engineering. Generally, to compare the performance of classical computing and quantum computing is not an easy task and several scenarios exist how a quantum computer could be utilized for the benefit of research or solving societal important problems which a classical computer alone could not solve.
A desirable scenario is one where problems that can not be solved by classical computers in any feasible amount of time, would be possibly solved with a quantum computer. This is one of the main goals of quantum computing, which is termed quantum supremacy. It is a demonstration that a programmable quantum device can solve a problem that any existing or future classical computer is not able to solve due to the problem's NP-hard complexity.
A second scenario is related to computational problems which could be solved more efficiently with both classical computers and quantum computers together, although not ruling out that a classical algorithm could solve the problem as efficienctly as the hybrid classical-quantum scheme when more powerful classical processing units would be available or a better classical algorithm is discovered.
The third category is comprised of problems that can not be solved more efficiently with a quantum computer. Thus, a careful analysis of the mathematical complexity of a mathematical problem formulation must be made to assess which computational system/architecture should be applied.
John Preskill published a paper titled "Quantum computing and the entanglement frontier" in November 2012 about quantum supremacy and the consequences it will have to several critical applications in society such as cryptography and optimization.
In October 2019, Google claimed with the Nature article "Quantum supremacy using a programmable superconducting processor" " to have achieved supremacy with a quantum processor called “Sycamore” to sample the output of a pseudo-random quantum circuit. They used 53 qubits to represent a dimensional state space.
Measurements from repeated quantum experiments sampled the resulting probability distribution, which then was verified using classical simulations. The process of sampling one instance of a quantum circuit a million times took about 200 seconds. With a classical computer that process would have taken 10,000 years. However, IBM stated with the blog post 'On "quantum supremacy"' that the computation of the Google experiment could be performed on a classical computer in 2.5 days. Many more quantum supremacy claims and demonstrations were made and the below table gives an overview of some of them.
Google (2019) | USTC (2020) | USTC (2021) | Xanadu (2022) | |
---|---|---|---|---|
Mathematical problem | Linear cross-entropy benchmarking fidelity of a pseudo-random quantum circuit | Gaussian Boson Sampling | Two-dimensional programmable quantum walks | Gaussian Boson Sampling |
Device | Superconducting transmon (Sycamore) | Optical interferometric network (Jiuzhang 2.0) | Superconducting transmon (Zuchonzhi 2.1) | Photonic processor (Borealis) |
Results/Claims | 200 seconds to sample one instance of a quantum circuit a million times in comparison to 200 million years classically | Measured sampling rate about 1014-fold faster than state-of-the-art classical method | Classical algorithm takes about 4.8x104 years while Zuchongzhi 2.1 takes about 4.2 hours | 9000 years for best classical algorithm to generate a single sample from programmed distribution, the quantum algorithm took 36 |
You can learn more about quantum computing for chemistry topics by signing up for the Dashboard Quantum or Machine Learning subscriptions.