Computational biology & systems immunology
immunoinformatics, machine learning, immune repertoires
Research aim
We integrate experimental and computational immunology to understand, predict, and engineer adaptive immune recognition. This includes developing software, profiling of immune responses, and designing computationally-guided therapeutics.
About us
Adaptive immune recognition plays a crucial role in both health and disease but remains challenging to study given the incredibly complex and personalized nature of the adaptive immune system. Recent advancements in systems immunology, bioinformatics, and artificial intelligence have revolutionized the resolution to which we can quantify adaptive immune responses. We are leveraging and developing technologies to computationally and functionally profile adaptive immune responses in pre-clinical and clinical settings spanning from infection, disease, and under homeostatic conditions. This is accompanied by the development of a comprehensive computational immunology ecosystem for immunogenomics data analysis that helps elucidate selection patterns of adaptive immune repertoires. Furthermore, the high-dimensional sequence space of B and T cell repertories provides feature-rich data that is well suited for machine learning and artificial intelligence algorithms. We therefore develop and implement novel computational and deep learning algorithms to learn representations of adaptive immune repertoires, to predict functional properties of B and T cell receptors, and to discover personalized immune signatures. Together, this fundamental and translational investigation into the language of the adaptive immune system holds the potential to improve the in silico design of adaptive immune therapeutics.