This research projec aims to contribute to the devel­op­ment of methods that extract infor­ma­tive and inter­pretable features from high-dimen­sional datasets, with a focus on uncov­er­ing high-level causally related factors that describe meaning­ful seman­tics of the data. This, in turn, can help us gain deeper insights into the repre­sen­ta­tions found within advanced gener­a­tive models, partic­u­larly founda­tion models and LLMs, with the goal of improv­ing their efficiency and safety. 

Amongst the leading causes of retinal dystro­phies world­wide is Retini­tis pigmen­tosa, a severe disease often caused by splice variants in the USH2A gene. This project aims to develop a safe CRISPR-based thera­peu­tic strat­egy for correc­tion of such splic­ing defects. Using enhanced-deletion nucle­ases, the disease-causing alter­ations can be elimi­nated, hereby restor­ing correct protein synthe­sis. The focus lies on safety features as well as the devel­op­ment of an inducible viral deliv­ery system for clini­cal application.

The research project inves­ti­gates the influ­ence of steri­cally demand­ing NHC-gold(I) complexes on the cycliza­tion of diyne deriv­a­tives. The focus is on the synthe­sis of various steri­cally demand­ing NHC-gold(I) complexes and their appli­ca­tion in diyne cycliza­tions, partic­u­larly examin­ing the reactiv­ity and selec­tiv­ity in gold-catalyzed reactions. Further inves­ti­ga­tions include theoret­i­cal calcu­la­tions and practi­cal appli­ca­tions of the synthe­sized cycliza­tion products for pharma­ceu­ti­cals or organic materials.

This project seeks to advance rare disease classi­fi­ca­tion using deep neural networks by address­ing key challenges such as limited data and high hetero­gene­ity. We will assess exist­ing models and their repre­sen­ta­tions, inves­ti­gat­ing how techni­cal varia­tions in medical data affect their characteristics. 

We aim to develop artifi­cial neural networks with brain-inspired circuits. Like in the brain, artifi­cial neurons and synapses are built with novel memris­tors, arranged in a cross­bar array. By combin­ing these with ultra-fast photon­ics, we seek to improve signal process­ing and matrix-vector multi­pli­ca­tions to address limita­tions from state-of-the-art archi­tec­tures. This approach aims to advance comput­ing solutions by reduc­ing energy consump­tion, compu­ta­tional time, and system complexity.

My doctoral project focuses on identi­fy­ing light-sensi­tive proteins called photore­cep­tors in the model organ­ism Chlamy­domonas reinhardtii. I aim to answer how these recep­tors regulate the inner biolog­i­cal clock known as the circa­dian rhythm. My major focus is to deter­mine the proper­ties of an unknown red light-sensi­tive photore­cep­tor and how this recep­tor regulates the clock. These insights can be used to under­stand how plants, in general, process light information.

This project focuses on the distri­b­u­tion of deep-sea macro­fauna (animals between 0.3 mm – 5 cm) in the deep Arctic Ocean across tempo­ral and spatial scales. I will test the hypoth­e­sis if environ­men­tal factors like ocean warming and ice retreat will affect commu­nity compo­si­tion. In addition, I will study the diver­sity, distri­b­u­tion, and connec­tiv­ity of isopods as they comprise an abundant and diverse group of the macro­fauna but are under­stud­ied in the Central Arctic. They are “brood­ers” – meaning they hatch their young in a brood pouch (imagine a tiny kanga­roo) and there­fore they usually do not disperse as far as animals with free-swimming larvae. 

3D print­ing at the nanoscale is an estab­lished technol­ogy. However, for certain appli­ca­tions it is still consid­ered prohib­i­tively slow. Conven­tion­ally, laser beam pulses illumi­nate one volume element after another in a light-sensi­tive ink, build­ing up the desired object. In this project, each laser pulse is holograph­i­cally shaped and illumi­nates a large number of voxels in paral­lel. This technique promises orders of magni­tude faster print­ing and shall be demon­strated for complex 3D structures.

Polar molecules enable the simula­tion of complex spin models and condensed matter phenom­ena due to their tunable long-range inter­ac­tions. This project aims to inves­ti­gate the transi­tion from a Bose-Einstein conden­sate of tetratomic molecules to a diatomic p‑wave super­fluid, referred to as BEC-BCS crossover. The super­fluid is of special inter­est as it is expected to host Majorana zero modes—quasi-particles ideal for fault-toler­ant topolog­i­cal quantum computing.

The project aims to accel­er­ate and improve the fabri­ca­tion of micro­ma­te­ri­als by 3D laser print­ing through the use of deep neural networks (DNNs). Physi­cal simula­tions of the print­ing process are devel­oped and used to train the DNNs. They can then, for example, charac­ter­ize the printed struc­tures already in the printer or pre-compen­sate objects in such a way that itera­tive charac­ter­i­za­tion and optimiza­tion outside the printer can be minimized.

The research project "Optical and electronic neuro­mor­phic systems" focuses on bio-inspired and resource-efficient concepts for neuro­mor­phic comput­ing. The aim is to realise these concepts in optical and electronic systems based on organic semicon­duc­tor materi­als and to describe their physi­cal foundations.

Human brains and vision-based robot­ics require inten­sive compu­ta­tion to recog­nize visual pattern features in various contexts and augmen­ta­tions, known as invari­ant pattern recog­ni­tion. The humming­bird hawkmoth (Macroglos­sum stellatarum) similarly uses pattern features on flowers to select suitable forag­ing sites, with only a fraction of the ‘compu­ta­tional power’. Aiming to under­stand how they do so with such efficiency, we will use behav­ioural, neural, and compu­ta­tional methods to uncover the algorith­mic basis of (invari­ant) pattern recog­ni­tion in insect pollinators.

In a space­time we have one time dimen­sions and multi­ple space dimen­sions. In our reality we experi­ence three space-like dimen­sions. Now in differ­en­tial geome­try, nothing keeps us from consid­er­ing manifolds with multi­ple time-like dimen­sions. In this project we study algebraic struc­tures, in partic­u­lar the group SO(p,q), which describe the dynam­ics and the geome­try of so-called pseudo-Riemann­ian hyper­bolic spaces with at least one time dimension.

Despite exten­sive research in person­al­ized medicine, promis­ing person­al­ized thera­pies still fail to trans­late into clini­cal practice. In my research project, I aim to construct a pathway model that predicts the effects of poten­tial thera­pies by combin­ing mecha­nis­tic model­ing and exper­i­men­tal approaches to meet ideal crite­ria for facil­i­tat­ing the trans­la­tion of research to patients.

Single magnetic molecules can be used as build­ing blocks to construct new artifi­cial spin systems which are inter­est­ing for future quantum devices. We use scanning tunnel­ing microscopy (STM) combined with electron spin resonance (ESR) to construct and inves­ti­gate such spin systems on a surface. This enables the study of funda­men­tal spin proper­ties on the atomic scale and explor­ing novel magnetic phenom­ena in multi-spin systems.

Collab­o­rat­ing with the LV Prasad Eye insti­tute, we inves­ti­gate sight recov­ery individ­u­als with a history of transient congen­i­tal blind­ness due to cataracts to unveil the neural mecha­nisms of sensi­tive periods in brain devel­op­ment. More specif­i­cally, we inves­ti­gate higher corti­cal repre­sen­ta­tions and whether and how they emerge if visual input arrives delayed e.g., not before mid-child­hood. The present PhD project will focus on object repre­sen­ta­tions and how they emerge in the inter­ac­tion with other visual areas. We expect a better under­stand­ing of how early experi­ence shapes adult brain connectivity.

Alzheimer’s disease (AD) has a multi­fac­to­r­ial etiol­ogy which includes, among others, vascu­lar dysfunc­tion and aberrant neuroim­mu­nity. We aim to inves­ti­gate the gene ABI3 as a poten­tial connec­tion between these two facets of AD patho­phys­i­ol­ogy. Through trans­genic murine models, and using a combi­na­tion of biochem­i­cal, immuno­his­to­chem­i­cal, and in vivo imaging techniques, we will explore how the late-onset AD risk variant S209F ABI3 affects neurode­gen­er­a­tion, immune fitness, and vascu­lar dynamics.

Rare genetic disor­ders lead to a failure to produce enough blood cells that are frequently fatal, seen most often among young children. These diseases are primar­ily monogenic, caused by the loss of function in a single gene. To inves­ti­gate the effects of this loss of function, my project seeks to mimic it outside of the human body, specif­i­cally in human bone marrow organoids (BMOs). By study­ing BMOs, the aim is to identify criti­cal factors contribut­ing to bone marrow failure and ultimately use this infor­ma­tion to develop new diagnos­tic methods.

Star clusters used to be consid­ered to consist of stars that all formed simul­ta­ne­ously and with the same elemen­tal abundances. The surpris­ing discov­ery that these clusters contain multi­ple popula­tions with charac­ter­is­tic abundance inhomo­geneities remains an enigma. I will inves­ti­gate whether rotational mixing is a plausi­ble culprit, using massive emission-line stars as tracers of rapid rotation. Also, I will assess the valid­ity of certain light elements as signa­tures of multi­ple populations.

DNA nanotubes are widely used as a mimic for cytoskele­tal filaments in bottom-up synthetic biology. Using a synthetic starPEG construct that acts as a crosslinker, we succeed in bundling the few nanome­ter thick DNA nanotubes. In bulk they self-assem­ble into micron-scale rings. We achieve their contrac­tion upon temper­a­ture increase or molec­u­lar deple­tion with crowing molecules such as dextran (in collab­o­ra­tion with Kierfeld group, TU Dortmund).

Galax­ies are embed­ded in a rich and complex atmos­phere – the circum­galac­tic medium (CGM). Under­stand­ing the processes going on in the CGM is inevitable for a self-consis­tent model for galaxy evolu­tion. Thus, we will shed some light on open questions about galaxy clusters using numer­i­cal simula­tions. We will analyze the already exist­ing cosmo­log­i­cal state-of-the art simula­tion Illus­trisTNG and also write new types of simulation.

Most animals exhibit bilat­eral symme­try, but asymmet­ric traits have repeat­edly evolved in differ­ent taxonomic groups. However, the genetic mecha­nisms respon­si­ble for asymmet­ric trait varia­tion remain unclear. We will use the scale-eating fish, Peris­so­dus microlepis, to dissect the genetic basis of its remark­able morpho­log­i­cal and behav­ioural asymme­try. This study will yield impor­tant insights into the mecha­nis­tic under­pin­nings of asymmet­ric devel­op­ment and the origin of evolu­tion­ary novelty.

Highly emotional memories are processed differ­ently from neutral ones. For negative experi­ences, this can result in maladap­tive memory forma­tion which may foster emotional psycho­log­i­cal disor­ders. This project aims to improve our under­stand­ing of adaptive and maladap­tive memory process­ing. We will analyze brain activ­ity in tasks that model maladap­tive memory symptoms. By this, we hope to identify entry points for treat­ments that counter­act maladap­tive memory formation.

Our study aims to analyse and map the cytoar­chi­tec­ture of the human hypothal­a­mus in histo­log­i­cal sections of 10 postmortem brains. As a result, we want to develop a high-resolu­tion 3D recon­structed histo­log­i­cal model of the hypothal­a­mus and its nuclei as a tool for assess­ing the struc­ture-function relation­ship and a proba­bilis­tic cytoar­chi­tec­tonic map of the hypothal­a­mus that will reflect the variabil­ity of hypothal­a­mic nuclei between individ­ual brains, in terms of size and location in standard refer­ence space.

The ability of non-resis­tant bacte­r­ial pathogens to survive antibi­otics during infec­tion (toler­ance) contributes not only to global rise of antibi­otic resis­tance, but also to chron­i­cal relapse of infec­tions. The aim of the project is to under­stand what contributes to bacte­r­ial revival after antibi­otic treat­ment and the under­ly­ing biolog­i­cal pathways. The findings of this project will contribute to better informed decisions on the selec­tion of antibi­otics to treat infec­tions and prevent relapse. 

Viruses are the most abundant biolog­i­cal entities on Earth. Although they infect members of the three domains of life, little is known about the infec­tion mecha­nisms of archaeal viruses. The aim of this research is to gain insight into the inter­ac­tion between halophilic archaeal cells and their viruses by using a combi­na­tion of light and electron microscopy with molec­u­lar biology and virolog­i­cal techniques.

Carbon nanotube (CNT) resonators will be designed and fabri­cated to exploit their sensing proper­ties. We will graft a single-molecule magnet (SMM) on such a CNT resonator in order to manip­u­late its spin states via the mechan­i­cal motion of the CNT. Using this nanome­chan­i­cal approach, single-molecule magnets will be inves­ti­gated with the long-term prospect of apply­ing them in future quantum technologies. 

Due to techno­log­i­cal advances we can now build devices that actively manip­u­late quantum mechan­i­cal objects, and using quantum objects as infor­ma­tion carri­ers has many impor­tant impli­ca­tions for future commu­ni­ca­tion. Quantum infor­ma­tion can be used to achieve perfect security and provide efficiency for commu­ni­ca­tion networks. This research project focuses on design­ing secure and anony­mous quantum commu­ni­ca­tion proto­cols in an effort to build a future quantum internet.

The detec­tion of gravi­ta­tional waves (GWs) has opened a new window on the universe, through which we can study the physics of black-hole and neutron-star mergers. By analyz­ing GWs we can infer proper­ties of the corre­spond­ing astro­phys­i­cal systems. Current analy­sis methods are however too compu­ta­tion­ally expen­sive to deal with the growing amount of data. My research is thus concerned with the devel­op­ment of more efficient methods for the GW analy­sis using power­ful machine learn­ing methods.

Land subsi­dence and ground­wa­ter salin­iza­tion are existence-threat­en­ing environ­men­tal changes in the Mekong Delta (MD). Their origin and process dynam­ics are not fully under­stood yet. By innov­a­tive measure­ment technol­ogy, field/lab inves­ti­ga­tions and numeric model­ing, a compre­hen­sive under­stand­ing of the processes is devel­oped and the effect of poten­tial counter­mea­sures can be examined. The elabo­rated knowl­edge is the basis for sustain­able water concepts in the MD and other delta areas worldwide.