Katarina Elez

/kah-tah-REE-nah EH-lez/

Berlin, Germany

I am a machine learning researcher in drug discovery with 6+ years of experience combining molecular machine learning and large-scale molecular simulations to design therapeutically relevant molecules.

I am currently pursuing a PhD in the AI4Science group at the Free University of Berlin and am a member of the IMPRS-BAC doctoral program at the Max Planck Institute for Molecular Genetics. I hold an MSc in Bioinformatics and a BSc in Computer Science, both awarded with honors.

During my PhD, I developed a virtual screening framework that identified a novel nanomolar TMPRSS2 inhibitor, which was experimentally validated against multiple coronaviruses and contributed to a patent application for a pharmaceutical composition to treat COVID-19. I have worked on hierarchical message passing neural networks for protein representation learning and am currently developing structure- and geometry-aware chemical language models. I also completed a research internship at Microsoft Research.

Outside research, I am passionate about teaching and scientific outreach. I co-founded Bioinformatika, a non-profit initiative dedicated to providing free bioinformatics education and mentoring opportunities to students from diverse socio-economic backgrounds.

I am motivated by challenging scientific problems with the potential for real-world impact. Ultimately, I hope to contribute to the development of AI technologies that make drug discovery more successful and more efficient.

news

No news so far...

selected publications

Simulations and active learning enable efficient identification of an experimentally-validated broad coronavirus inhibitor

Katarina Elez, Tim Hempel, Jonathan H. Shrimp, Nicole Moor, Lluís Raich, and 7 more authors

Nature Communications, 2025

Abs Bib HTML Code

Drug screening resembles finding a needle in a haystack: identifying a few effective inhibitors from a large pool of potential drugs. Large experimental screens are expensive and time-consuming, while virtual screening trades off computational efficiency and experimental correlation. Here we develop a framework that combines molecular dynamics (MD) simulations with active learning. Two components drastically reduce the number of candidates needing experimental testing to less than 20: (1) a target-specific score that evaluates target inhibition and (2) extensive MD simulations to generate a receptor ensemble. The active learning approach reduces the number of compounds requiring experimental testing to less than 10 and cuts computational costs by ∼29-fold. Using this framework, we discovered BMS-262084 as a potent inhibitor of TMPRSS2 (IC50 = 1.82 nM). Cell-based experiments confirmed BMS-262084’s efficacy in blocking entry of various SARS-CoV-2 variants and other coronaviruses. The identified inhibitor holds promise for treating viral and other diseases involving TMPRSS2.
@article{elez_simulations_2025, title = {Simulations and active learning enable efficient identification of an experimentally-validated broad coronavirus inhibitor}, author = {Elez, Katarina and Hempel, Tim and Shrimp, Jonathan H. and Moor, Nicole and Raich, Lluís and Rocha, Cheila and Winter, Robin and Le, Tuan and P\"{o}hlmann, Stefan and Hoffmann, Markus and Hall, Matthew D. and Noé, Frank}, year = {2025}, journal = {Nature Communications}, volume = {16}, pages = {6949}, doi = {10.1038/s41467-025-62139-5}, altmetric = true, dimensions = true, bibtex_show = true, selected = true }
Synergistic Inhibition of SARS-CoV-2 Cell Entry by Otamixaban and Covalent Protease Inhibitors: Pre-Clinical Assessment of Pharmacological and Molecular Properties

Tim Hempel, Katarina Elez, Nadine Krüger, Lluís Raich, Jonathan H. Shrimp, and 8 more authors

Chemical Science, 2021

Abs Bib HTML

SARS-CoV-2, the cause of the COVID-19 pandemic, exploits host cell proteins for viral entry into human lung cells. One of them, the protease TMPRSS2, is required to activate the viral spike protein (S). Even though two inhibitors, camostat and nafamostat, are known to inhibit TMPRSS2 and block cell entry of SARS-CoV-2, finding further potent therapeutic options is still an important task. In this study, we report that a late-stage drug candidate, otamixaban, inhibits SARS-CoV-2 cell entry. We show that otamixaban suppresses TMPRSS2 activity and SARS-CoV-2 infection of a human lung cell line, although with lower potency than camostat or nafamostat. In contrast, otamixaban inhibits SARS-CoV-2 infection of precision cut lung slices with the same potency as camostat. Furthermore, we report that otamixaban’s potency can be significantly enhanced by (sub-) nanomolar nafamostat or camostat supplementation. Dominant molecular TMPRSS2-otamixaban interactions are assessed by extensive 109 μs of atomistic molecular dynamics simulations. Our findings suggest that combinations of otamixaban with supplemental camostat or nafamostat are a promising option for the treatment of COVID-19.
@article{hempel_synergistic_2021, title = {Synergistic Inhibition of {{SARS-CoV-2}} Cell Entry by Otamixaban and Covalent Protease Inhibitors: Pre-Clinical Assessment of Pharmacological and Molecular Properties}, author = {Hempel, Tim and Elez, Katarina and Kr{\"u}ger, Nadine and Raich, Llu{\'i}s and H.~Shrimp, Jonathan and Danov, Olga and Jonigk, Danny and Braun, Armin and Shen, Min and D.~Hall, Matthew and P{\"o}hlmann, Stefan and Hoffmann, Markus and No{\'e}, Frank}, year = {2021}, journal = {Chemical Science}, volume = {12}, number = {38}, pages = {12600--12609}, publisher = {Royal Society of Chemistry}, doi = {10.1039/D1SC01494C}, altmetric = true, dimensions = true, bibtex_show = true, selected = true }
Distinguishing Crystallographic from Biological Interfaces in Protein Complexes: Role of Intermolecular Contacts and Energetics for Classification

Katarina Elez, Alexandre M. J. J. Bonvin, and Anna Vangone

BMC Bioinformatics, 2018

Abs Bib HTML Code

Study of macromolecular assemblies is fundamental to understand functions in cells. X-ray crystallography is the most common technique to solve their 3D structure at atomic resolution. In a crystal, however, both biologically-relevant interfaces and non-specific interfaces resulting from crystallographic packing are observed. Due to the complexity of the biological assemblies currently tackled, classifying those interfaces, i.e. distinguishing biological from crystal lattice interfaces, is not trivial and often prone to errors. In this context, analyzing the physico-chemical characteristics of biological/crystal interfaces can help researchers identify possible features that distinguish them and gain a better understanding of the systems.
@article{elez_distinguishing_2018, title = {Distinguishing Crystallographic from Biological Interfaces in Protein Complexes: Role of Intermolecular Contacts and Energetics for Classification}, shorttitle = {Distinguishing Crystallographic from Biological Interfaces in Protein Complexes}, author = {Elez, Katarina and Bonvin, Alexandre M. J. J. and Vangone, Anna}, year = {2018}, journal = {BMC Bioinformatics}, volume = {19}, number = {15}, pages = {438}, doi = {10.1186/s12859-018-2414-9}, altmetric = true, dimensions = true, bibtex_show = true, selected = true }