Seeing in atomic details: from molecules to functional cells
Professor Alan Mark studies how biological systems function in atomic detail, from how cell membranes form to how viruses penetrate them and drugs affect them. Since it is not possible to directly observe the properties of individual molecules as they interact with cellular structures, he uses theoretical approaches and computer simulation techniques to build detailed computational models that can predict the properties of real systems.
Alan has always wanted to know how living systems work, in particular how different components within cells can spontaneously self-assemble into functional structures. Understanding that complexity means delving into more and more detail, right down to individual atomic interactions.
He studied chemistry and biochemistry before looking at how insulin binds to receptors for his PhD research. He then began working to determine the 3D shapes of proteins using nuclear magnetic resonance imaging. This led him to move to the Netherlands and then Switzerland to join one of the world-leading groups developing molecular simulation techniques to model the key elements of cellular function, such as how proteins fold. Since then he’s used computer modelling to explore other critical cellular processes such as how receptors on the surface of a cell recognise specific drugs and how protein misfolding can lead to disease.
What drew him to science?
“I was always interested in science,” says Alan. “How do things work? How do leaf cells respond to some combination of temperature and day length to recognise when it’s time to fall off a tree in autumn? How do molecules get in and out of cells? It’s all just interactions between atoms and molecules. If we can model these interactions correctly, we can understand all aspects of a system. We’ve progressed a lot but are still incredibly naïve about how biological systems really work at an atomic level.”
Research with supercomputers
Molecular dynamics simulation techniques involve calculating the interactions between every pair of atoms in a particular system. Once all the forces acting on each atom are known the system can be propagated step-wise through time using Newton’s laws of motion. Millions of these small steps in time are combined to make a molecular ‘movie’ of the system. These movies allow us to see how molecules interact and move, how they find each other and assemble into a working structure.
“The challenge is to describe the interactions between atoms with enough accuracy to be able to examine the properties of interest, and to reproduce measurable properties of real systems,” explains Alan. “Biomolecular systems are large and complex, involving up to millions of atoms, and must be modelled over significant timescales to explore all of the possible interactions. It’s so computationally intensive it’s only possible using a supercomputer.”
Alan has spent his career improving the reliability of these models, and validating them wherever possible against experimental data so that their predictions can be used with confidence.
Real world solutions
Alan has used molecular dynamics simulation techniques to demonstrate how membranes self-assemble, how different lipids (the oils that make up cell membranes) contribute to antimicrobial resistance, how receptors get triggered by drugs and how peptides bind to membranes. He and his team have also created a globally-recognised molecular modelling tool and database, the Automated Topology Builder. This is used by researchers around the world to generate atomic interaction parameters for novel molecules as required for computational drug design and the design of new materials.
Alan has previously used Pawsey supercomputing to simulate the fusion-active forms of the Ebola virus and the Hendra virus to study exactly how they infect cells, as well as inform drug design efforts to block their formation. He is currently determining the structure of the fusion-active intermediate of the SARS-CoV-2 protein spike as a potential target for drug design. His team is also working to ensure all researchers wanting to test how existing drugs may be repurposed to treat COVID-19 have ready access to the highest quality interaction parameters possible.