The human adaptive immune system is a natural diagnostic and therapeutic. Disease diagnostic information and therapeutic potential is directly encoded into antibody and T-cell immune repertoires. The advent of high-throughput sequencing has enabled an unprecedented accumulation of big immune repertoire sequencing data (immunegenotype). However, as of yet, we lack the computational and experimental methods that help us decode the immune grammar that translates immunegenotype to immune state diagnosis and prediction of antigen binding (immunephenotype). We believe that learning to read and write the immune repertoire language is key for the development of entirely novel, nature-inspired precision medicine immunodiagnostics and immunotherapeutics.

Our projects are grouped into three main areas of research and development:

ai_researchArtificial intelligence methods for deciphering the human immune repertoire language. Deciphering the immune information (immune status, antigen binding) encoded into antibody and T-cell repertoires is of paramount importance for the development of vaccines, diagnostics and therapeutics and requires machine learning approaches (artificial intelligence).  We focus on developing deep learning approaches based on CNN, RNN, and autoencoders in order to learn how to read and write the immune repertoire language. We specifically focus on cracking the blackbox of deep learning by aiming to understand which parts, or patterns, of the immune repertoire or immune receptor sequence contain the highest immunological information. We hypothesize that these sequence patterns form an immune grammar that may used for constructing entirely novel and therapeutically relevant immune receptor sequences.


ssms_researchSingle-cell immune repertoire and immuno-mass spectrometry methodsHigh-throughput sequencing has enabled unprecedented insight into the diversity and complexity of immune receptor repertoires. The advent of single-cell sequencing has recently provided the opportunity to further increase the resolution of immune repertoire analyses. As of yet however, the link between genomic antigen receptor diversity and the effector phenotypic serum antibody response has remained mostly unclear rendering the understanding of vaccination and disease-specific responses unfeasible.  Therefore, our research focuses on developing experimental platforms that link droplet-based high-throughput single-cell sequencing with the large-scale mass-spectrometry analysis of serum antibody repertoires in order to resolve the diversity and specificity of the effector serum antibody repertoire at single antibody resolution.


network_researchAsset 4New bioinformatics platforms for high-dimensional immune repertoire analysis. The integration of high-throughput repertoire sequencing data for easier downstream systems immunology analyses has remained a challenge so far. We focus on developing novel bioinformatic tools that enable multi-source (antibody genomics, proteomics, binding assays) data integration in order to facilitate large-scale computational analyses. Specifically, besides machine learning, we work on using network, entropy and structural analytical methods for deconvolving the complexity of antibody repertoires and understanding the underlying structure of adaptive immunity.


We apply the developed methods to decipher the specific immune repertoire architecture of  a wide array of human diseases spanning autoimmune disease (RA, T1D, Celiac disease), infection (Tuberculosis, Influenza) and cancer (GBM).

Our work is funded by the University of Oslo, UiO:LifeScience, Erasmus+, Horizon2020, Helmsley Charitable Trust.