Marius Wanko Email and Phone Number

Awarded with a 3-year national research grant (1st place in chemistry), I started as a postdoctoral researcher at UPV/EHU. I used this opportunity to form my own group, supervising a PhD student and a postdoc, working on switchable and far-red emitting fluorescent proteins. Our ambitious goal was to simulate the light-induced processes in these proteins to guide the design of novel, genetically engineered fluorescent proteins for biomolecular applications, which would be produced and tested by the Nienhaus group at KIT. The project turned out to be far more challenging than anticipated from previous work due instabilities in the chromophore binding pocket and mutants that would not mature in experiment. Both issues showed the limitations of state-of-the-art biophysical models and could possibly be overcome by future machine-learning based models, like AlphaFold and RoseTTAfold, that distill the folding behavior of millions of proteins. As a consequence, I took a statistical approach to explore the range of photophysical properties of these proteins and started my journey into data science and machine learning.

As a Consolider at Basque research institute nanoGUNE, I fully focused on my research. While continuing collaborations with Braunschweig and Karlsruhe (KIT), I initiated a very fruitful collaboration with experimental groups at the Danish ELISA ion storage ring for molecular action spectroscopy, helping them to interpret experimental data and study microsolvation. I also trained local groups at nanoGUNE and University of the Basque Country in quantum chemical calculations. I prepared my PhD defense and a 3-year research proposal on fluorescent proteins.

At the University of Bremen, I helped to build up the Bremen Center for Computational Materials Science (BCCMS), founded by my PhD supervisor Thomas Frauenheim. At the same time, I continued my research on retinal proteins, developing strategies for precise theoretical spectroscopy on molecules interacting with a solvent, protein, or solid-state environment, at the Technical University of Braunschweig, in the group of my second supervisor, Marcus Elstner (now KIT).

During my PhD, I developed models to simulate the first step of the vision process, the excited-state dynamics of the retinal molecule (chromophore) in rhodopsin proteins. I also studied the color tuning of such chromophores by the protein, which allows color vision in humans and animals, and light-harvesting and sensing in archaebacteria.With my results, I supported two 6-year projects within DFG research group FOR 490 "Molecular Mechanisms of Retinal Protein Action: A Combination of Theoretical Approaches"