Over the last decade, the field of immunology has experienced a true revolution. Where once immunology was a largely descriptive field of research, immunologists nowadays generate large, quantitative, and temporal data sets on gene and protein expression, diversity of the lymphocyte repertoire, and the coevolution between pathogens and the immune system. Immunology has thereby become a lot more quantitative. Such quantitative insights are extremely valuable to understand the complexity of the immune system in health and disease. The interpretation of complex and quantitative data sets requires computational approaches and specific expertise.

The Computational Immunology Core of the Center for Translational Immunology at UMC Utrecht brings together bioinformaticians, biostatisticians and mathematical modelers, all working on immunological questions. They help interpret the large, complex, and expensive, high throughput datasets that are being generated by immunologists. Because innovations in big data generation occur at an extremely rapid pace, the Computational Immunology Core (with prof. José Borghans PhD, Julia Drylewicz PhD and Alexander Yermanos PhD as principal investigators) also works on continuous development of new methodologies to interpret the data, including artificial intelligence (AI) methods. In addition, by working in very close collaboration with experimental immunologists and clinicians, they form an important bridge between the wet and dry lab as well as the clinic. By being involved in projects from the start, they help to optimize the study design and make sure that the optimal analyses are performed.

Systems immunology approach

The growing body of big data is transforming immunological research and is already impacting patient care. All the layers of information that can nowadays be retrieved from patient cells, such as gene expression or repertoire data, open up the possibility to discover differences between patients with the very same diagnosis, for whom different treatments may be optimal. For example, patients with atopic dermatitis, a complex and highly heterogeneous inflammatory skin disease, turn out to fall in different groups of patients (called endotypes), which each may require a different therapy. Translational immunology is thereby shifting from a “one treatment fits all” to a more personalized approach. On the flipside, symptoms that are classically defined as originating from different diseases may in fact have common underlying immune mechanisms, and may thus benefit from similar treatment approaches. A systems immunology approach could thus also lead to a reclassification of disease based on immune profiles rather than disease symptoms. Such reclassifications will help to develop new drugs and to reassign already existing drugs to other diseases.

Next generation of immunologists

The increased need for computational expertise in immunological research also means that we need to prepare the next generation of immunologists. We need to equip students and employees alike with the required expertise to analyze their own high-throughput data and to interpret and judge computational work in the scientific literature. In order to reach this goal, we have developed the Infection & Immunity Master’s course “Computational Immunology”, which exposes students to the use of biostatistics, bioinformatics and mathematical modeling in immunology. Besides, the Computational Immunology Core provides training and guidance for PhD students and post-docs within the Center for Translational Immunology, both for people doing purely computational projects and for people with an experimental or clinical background doing projects with a dry-lab component. The need for this training is beyond dispute, because we foresee that computational work will play an ever-increasing role in Translational Immunology.