Meet the Expert: Professor Flemming Hansen

31 January 2024

Flemming Hansen received his PhD in biophysical chemistry in 2005. Following postdoctoral training, he joined UCL in 2010 as a David Phillips Fellow. Elected as a Royal Society of Chemistry Fellow in 2015, he became a full professor and Chair of NMR spectroscopy at UCL in 2017.

Can you explain, in simple terms, how the ‘molecular dance’ of a protein correlates to its biological function?

Answer: Any machine, which we come across in our daily life, needs some of its components to move in order to work. For example, all the components of a clock need to turn and move in a coordinated manner for the clock to work. It is the same with macromolecules and enzymes, which are small machines in the human body. My group currently researches a hydrolase enzyme, which cuts chemical bonds within other proteins. As the two blades of a pair of scissors need to move in order to cut a string, the hydrolase enzyme needs to move its various parts, or dance, in a specific way to function. When this molecular dance is out of harmony the enzyme will not work properly and diseases can develop.

What drew you to the study of human histone deacetylases (HDACs) and can you explain to a non-expert what these are?

Answer: Histone deacetylases are part of a group of enzymes named hydrolases; they cut chemical bonds within other molecules in the cell. Specifically, HDACs remove acetyl groups from proteins that organise DNA, thereby preventing the cell’s machinery to access that particular part of the DNA. You could say that HDACs make parts of the genetic code silent.

HDACs simply have everything I am interested in. They clearly need to be dynamic and flexible to function, and they are involved in various diseased states, including cancers and neurological disorders. This means that what we learn about the function of HDACs potentially can guide the development of new treatments.

How does NMR help you in characterising macromolecular motions, interactions, and dynamics?

Answer: NMR spectroscopy is a unique experimental biophysical technique because it provides insight into the motions and dynamics of enzymes with atomic resolution. Moreover, because of the flexibility of NMR, we can design tailored methods to characterise the macromolecular motions on time scales that coincide with the time scales of the functions. In other words, whereas many techniques provide pictures of macromolecules, NMR can provide the experimental input to make a movie of the macromolecule of interest. As with the pair of scissors, if an alien was shown a static picture of scissors, they would most likely not understand how it functions, however, if someone showed them a movie it would be very clear.

What are the current challenges or limitations in using NMR for studying macromolecular dynamics, and how are you addressing them?

Answer: A plethora of NMR methods have been developed over the last four decades, and as mentioned above, we now have methods available to characterise nearly every aspect of macromolecular motions. However, the analysis of NMR data is often challenging and sometimes requires decades of training. This, sadly, means that only a fraction of scientists effectively have access to the full toolbox of NMR methods. We are currently developing autonomous AI tools to analyse even the most complex of NMR data, which means that soon all scientists will be able to implement NMR into their research programmes.

You have published several papers on integrating Deep Neural Networks (DNNs) for NMR data analysis and interpretation. How do you see machine learning further impacting the field of NMR spectroscopy and macromolecular dynamics?

Answer: Although I think we still have some research to do, it is conceivable that in the near future, AI agents will autonomously operate NMR machines, akin to how self-driving cars already navigate busy roads with minimal input from passengers. For example, a biochemist will be able to provide the AI-linked NMR machine with simple directions or questions about macromolecules, and the results will be generated by the AI agent operating the NMR machine. This development will unlock the full potential of NMR spectroscopy, removing the barrier of decades of training that currently limit NMR’s accessibility. Once everyone has access to the powers of NMR, I think we will see very impressive developments and applications beyond what we currently can imagine.

What advice do you have for young scientists interested in the field of macromolecular dynamics?

Answer: First and foremost, they will need to be curious and have a strong desire to understand the intricacies of macromolecular function. A strong foundation in math and physics will help, but generally, they should love what they do, be creative, and think big.

What future research projects are you particularly excited about?

Answer: I am very excited about various aspects of integrating AI with NMR spectroscopy and other areas of magnetic resonance. I am currently excited to broaden my research to integrate AI with magnetic resonance imaging (MRI); a technique most people know from the hospital and a technique which is based on the same 1938 quantum physics discovery as NMR. With the development of portable MRI devices, I imagine that AI will make it possible to make critical decisions about injuries much more quickly.