Nikos Hatzakis

From Vision to Reality: Direct real time imaging biological events in real time and augmented by machine learning analysis
Nikos Hatzakis

Professor University of Copenhagen, Department of Chemistry, Faculty of Science, University of Copenhagen
Visiting Professor Harvard Medical school , department of pediatrics Boston Children hospital
Director of Novonordisk center for Optimised oligo escape and Control of Disease (COE)
Director of Novonordisk center for 4D cellular dynamics

Current advances in imaging technologies compounded with recent advances in sophisticated, machine learning drivel analysis, have propelled us into an era where the direct observation of biological phenomena in real time is not a phantasy—it is now a reality. These innovations at the bleeding edge of science are transforming our understanding of biological processes, once confined to the realm of imagination, into actionable insights.

I will present recent breakthroughs of my lab pushing the boundaries of biological imaging and analysis. These advances enable us to directly observe biological processes as they occur, extract quantitative mechanistic information, and leverage this knowledge to control aberrant biological functions.

Specifically I will discuss our work on the real-time observation of protein aggregation, cell entry and delivery pathways as well as machine learning analysis, I will focus on

  1. Live-Cell imaging and 4D Tracking: Approaches to track hundreds of individual nanoparticles (viruses LNPs, protein aggregates), allowing direct visualization of genetic material release and downstream expression events.
  2. Machine Learning Breakthroughs: Deep learning approaches accelerating analysis by six orders of magnitude, achieving up to 95% accuracy in milliseconds (instead of days) to pinpoint critical timepoints in these processes. In essence offering live analysis

Relevant publications from my lab

  1. Jacob Kaestel- et al. Deep learning assisted single particle tracking for automated correlation between diffusion and function, accepted in a Nature Methods 2025
  2. Bender, S. W. B., Dreisler, M. W., Zhang, M., Kæstel-Hansen, J. & Hatzakis, N. S. SEMORE: SEgmentation and MORphological fingErprinting by machine learning automates super-resolution data analysis. Nat. Commun. 15, 1763 (2024).
  3. Stella, S. et al. Conformational Activation Promotes CRISPR-Cas12a Catalysis and Resetting of the Endonuclease Activity. Cell 175, 1856-1871.e1821 (2018).
  4. Malle, M. G. et al. Single-particle combinatorial multiplexed liposome fusion mediated by DNA. Nature. Chemistry. (2023).
  5. Pinholt, H. D., Bohr, S. S.-R., Iversen, J. F., Boomsma, W. & Hatzakis, N. S. Single-particle diffusional fingerprinting: A machine-learning framework for quantitative analysis of heterogeneous diffusion. Proceedings of the National Academy of Sciences 118, e2104624118 (2021).
  6. Jensen, S. B. et al. Biased cytochrome P450-mediated metabolism via small-molecule ligands binding P450 oxidoreductase. Nat. Commun. 12, 2260 (2021).
Scroll to Top