This project focuses on characterizing the dynamic processes that drive acute leukemia cell evolution, including phenotypic drift and lineage switching, by integrating multimodal single-cell data with additional topological and kinetic measures (e.g., RNA velocity). 

The heterogeneity and dynamic behavior of leukemic blasts present significant challenges for minimal residual disease (MRD) monitoring and the assessment of treatment responses. Current single-cell methodologies provide only static snapshots of leukemic cells at specific time points; however, computational approaches can infer the underlying dynamics from these static views. This project focuses on characterizing the dynamic processes that drive acute leukemia cell evolution, including phenotypic drift and lineage switching, by integrating multimodal single-cell data with additional topological and kinetic measures (e.g., RNA velocity). The project employs advanced computational and mathematical methods—including computational topology, stochastic modeling, and deep learning—to dissect the state transitions and fate-determining processes of leukemic cells. A core objective is to develop a computational framework capable of integrating diverse data modalities and analyzing the associated mathematical structures. These techniques will identify therapy-resistant cell subsets and link MRD findings to their developmental context.

The analytical framework will be applied to MRD samples from leukemia patients exhibiting phenotypic changes, lineage switching, or delayed therapeutic responses. The research is expected to provide a detailed understanding of the mechanisms driving phenotypic transitions in residual leukemic blasts and to identify key features of drug-resistant leukemic clones. Additionally, it will deliver a broadly applicable computational framework for high-dimensional single-cell data analysis, specifically designed to study MRD dynamics and the complex interplay of cells within their tissue environment.

Host:

Charles University, Prague, Czech Republic

Supervisors:

Jan Trka and Dr. Jan Stuchly 

Profile Docatoral Candidate:

Master’s degree in mathematics, physics, (bio)informatics, life sciences, other (bio)medical sciences, or related and proficiency in programming and bioinformatics, and/or artificial intelligence.

Duration:

36 months, and extendable up to 48 months under the standard terms of the Ph.D. candidate position at the host institution. Starting 1-10-2025. 

In addition:

As part of the project, the doctoral candidate will intern at the VIB-KU Leuven Center for Cancer Biology (1 month, Belgium, supervisor Jan Cools) to learn single cell sequencing technologies and at ORTEC B.V. (2 months, Netherlands, supervisor John Jacobs) to obtain experience with clinical artificial intelligence.