ERC Starting Grant for TU Delft researchers

News - 03 September 2020 - Communication

The ERC has awarded its 2020 Starting Grants to early-career researchers. Two of them are scientists from TU Delft. This European funding will help individual scientists and scholars to build their own teams and conduct pioneering research.  

The ERC Starting Grant winners from TU Delft are:

Caroline Paul

BioAlk - Sustainable production of organic molecules

In recent years, industry has increasingly been using Nature’s catalysts, enzymes, for the production of pharmaceuticals, aromatics and other useful molecules. This embrace of organic chemistry is quite understandable. After all, enzymes can be more sustainable and cheaper than the toxic substances that are still widely used to drive chemical reactions.

Unfortunately, few enzymes have been developed so far to efficiently form so-called 'carbon-carbon bonds' - the molecular connections that form the basis of all organic molecules. At present, organic molecules are therefore still made through various classic chemical methods, such as using rare metals like palladium. Caroline Paul wants to change that. Her ERC project, called BioAlk, aims to adapt existing enzymes to allow them to create carbon-carbon bonds. Inserting the gene responsible for the production of such an enzyme in E.coli bacteria would allow it to be obtained in large quantities. 

The time is ripe for this project, believes Paul. "Nowadays, we can very precisely modify single amino acids to make mutants of a particular enzyme. The aim is that such an engineered enzyme will then do the chemistry we want. We are getting better and better at creating useful mutants, with the help of bioinformatics we can predict the effect of specific mutations to some extent". This scientific discipline, the artificial improvement of enzymes, is very active. In 2018, the American researcher Frances Arnold was awarded the Nobel Prize in Chemistry for her pioneering work in this field.

Caroline Paul is optimistic about the chances of her project becoming a success: "We will be able to achieve proofs of concept, which will open new avenues for chemical synthesis." The evaluators of the project were very enthusiastic about Paul’s ambition. One of them wrote that the project has the potential to be a real game changer, which could give Europe a leading position in the field of biocatalysis.

Dr. Caroline E. Paul

Assistant Professor


Peyman Mohajerin Esfahani

Recent developments in sensing and communication technology offer unprecedented opportunities by ubiquitously collecting data at high detail and at large scale. Utilization of data at these scales, however, poses a major challenge for control systems, particularly in view of the additional inherent uncertainty that data-driven control signals introduce to systems behaviour. This effect has not been well understood to this date, primarily due to the missing link between data analytics techniques in machine learning and the underlying physics of dynamical systems.

Esfahani addresses this issue by proposing a novel control design paradigm embracing ideas from the emerging field of distributionally robust optimization (DRO). DRO is a decision-making model whose solutions are optimized against all distributions consistent with given prior information. Recent breakthrough work, among others by the PI of this proposal, has shown that many DRO models can be solved in polynomial time even when the corresponding stochastic models are intractable. DRO models also offer a more realistic account of uncertainty and mitigate the infamous post-decision disappointment of stochastic models.   

Peyman Mohajerin Esfahani: “This proposal lays the theoretical foundation for distributionally robust control and aims to make progress along four directions. (i) Decision-dependent ambiguity: I introduce the concept of invariant ambiguity sets to encompass the dynamic evolution of uncertainty. (ii) Dynamic programming: I establish a dynamic programming characterization of the proposed DRO models and provide tractable approximation schemes along with rigorous theoretical bounds. (iii) Safe and memory-efficient learning: Leveraging modern tools from kernel methods and online-optimization, I propose tractable, yet provably reliable, synthesis tools. (iv) I plan to develop a tailor-made modelling language and open-source software to make distributionally robust control methods accessible in industry-size applications.”