Benjamin Schubert

Already early on in my academic career, I was fascinated by the complexity of the immune system and started applying machine learning methods to model and predict specific aspects of the adaptive immune response during my Bachelors in Bioinformatics at the University of Tübingen. Later in my Masters, I developed an benchmarked computational tools for computer-aided vaccine design as a DAAD Full-time Graduate Fellow at the Center of Biological Sequence Analysis, Technical University of Denmark, under the supervision of Ole Lund and Morten Nielsen.

During my Ph.D. at University Tübingen in the group of Oliver Kohlbacher, I mostly focused on personalized cancer vaccine design; specifically, on how to optimally select and assemble neoepitopes into minigene designs that can be administered as RNA or polypeptide vaccines. In a recent collaborative effort, we could demonstrate that such algorithmically designed vaccines, indeed positively impact vaccine efficacy by increasing the likelihood of recovering the selected neoepitopes during antigen processing and often outperform human-designed minigene vaccines.

For my postdoctoral training at Harvard Medical School and Dana-Faber in the labs of Debora Marks and Chris Sander, I switch gears and developed statistical models of proteins and genomes to learn genotype-phenotype interactions and used these models for protein re-engineering to reduce side effects of modern biotherapeutics. To this end, I developed approaches that successfully combined machine learning methods and combinatorial optimization algorithms.

Since 2018, I joined the Institute of Computational Biology, Helmholtz Center Munich as a Team Leader, where my team and I have started to leverage newly arising single-cell immune profiling experiments to learn more about the rules that guide T-cell activation and to build deep learning models to predict T-cell specificity. This knowledge can then be used to unravel the disease etiology of autoimmune diseases and how we can design T-cell-based immunotherapies with high specificity and avidity.