Gareth Conduit

Gareth Conduit

Royal Society University Research Fellow in the Theory of Condensed Matter Group, at the University of Cambridge. The group develops and applies machine learning to design new materials & drugs and researches quantum phenomena. The machine learning technology Alchemite™ is commercialized for materials and industrial chemicals design through Intellegens, and as the drug design suite Cerella™ by Optibrium. Further details can be found in the research stories below, news coverage, recent talks, and publications.

Material design

Through the stone, bronze, and iron ages the discovery of materials has chronicled human history. The coming of each age was sparked by the chance discovery of a new metal. Despite the central importance of materials in enabling breakthrough technologies, even today the only way to develop new materials is through experimental driven trial and improvement. We proposed, implemented, and employed a machine learning tool, Alchemite™, that has specialist capabilities to handle the typical sparse experimental data set. The approach is generic so we have applied it to both industrial materials and drug design. Alchemite™ is now being commercialized by Intellegens.

In a collaboration with Rolls Royce the tool designed four alloy families. The alloys include two nickel-based alloys for jet engines (patents EP14157622, US2013/0052077A2), and two molybdenum alloys for forging hammers (patents EP14153898, US2014/177578, EP14161255, US2014/223465), and an alloy for 3D printing of combustors. Each alloy has thirteen individual physical properties that are predicted to match or exceed commercially available alternatives, and for each alloy eight properties have been experimentally verified.

Alchemite™ can also juxtapose experimental data with first principles computer simulations, demonstrated to explore new lubricants, metal organic frameworks, and battery lifetime predictions. The methodology can be used not only for materials discovery, but also for imputing and finding errors in databases. In a project with ANSYS Granta, Alchemite™ uncovered over a hundred errors in commercial alloy and polymer databases.

Drug design

Designing drugs is a huge computational challenge: finding the drug molecule that will correctly affect the behavior of all 10,000 proteins in the human body. The search is complicated by the dearth of information: just 0.05% of the known drug molecule-protein activity levels are actually known. The machine learning tool, Alchemite™, developed for materials design is perfectly suited for handling sparse data, therefore it is now being used in several academic projects and is marketed as the Cerella™ suite by Optibrium.

The additional insights offered by Alchemite™ has reduced the cost of performing additional experiments and accelerated the discovery of drugs. Alchemite™ achieves the best prediction accuracy ever seen on a benchmarking data set published by Novartis, and won the Open Source Malaria competition with the only molecule experimentally verified to be active. Alchemite™ is being used by multinational pharmaceutical companies to accelerate their drug discovery programs.

Magnetic spin spiral

All of the electrons in a materials solid interact with each other, so that when one electron moves it pushes all of the other electrons in the material. Furthermore, electrons are quantum particles so they obey the counter-intuitive laws of quantum mechanics. The juxtaposition of many-body interactions and quantum mechanics leads to exotic phenomena including ferromagnetism and quantum mechanics.

An electron gas with contact repulsive interactions is a deceptively simple interacting system, yet it displays a remarkably rich range of phenomena. At mean-field level the electron gas was predicted by Stoner to undergo a mean-field transition into an itinerant ferromagnet. This phase has never been cleanly observed in the solid state, however our study underpinned the first experimental exploration of its properties in an ultracold atomic gas. Moreover, quantum fluctuations mean that a variety of other inhomogeneous magnetic states are energetically favorable, with our suggestion of a spin spiral state being first observed in CeFePO in 2012.

Few trapped atoms

A few-fermion ultracold atom system presents an alternative arena to study strongly interacting fermions. The system allows strong correlations to be probed and explained within an exactly solvable and experimentally measurable system. Our studies of the consequences of repulsive and attractive interactions in this system was followed by the experimental realization and characterization of a few-atom Fermi sea by the Jochim group. The system shows that the crossover to many-body physics takes place with just six fermions, making this system an ideal playground to develop the intuition and understanding of many-body physics.