NextGen Immunology: Modeling microbial HLA-II peptidomes and identification of disease-relevant T cell epitopes using artificial intelligence and immunopeptidomics

Doctoral Researcher

Associated Doctoral Researcher

Associated Principal Investigator

Background and current state of research

In 2020, doctoral researchers from at least two or more RTG projects working in different disciplines could apply for extra funding for a small interdiciplinary project within the RTG.

The aim of this offer is

  • to enhance the scientific collaboration within the RTG,
  • to widen interdisciplinary skills of the doctoral researchers and
  • to provide the doctoral researchers the opportunity to practice
    • writing a proposal,
    • working as a team,
    • coordinating a small own research project (one of the junior researchers helds the position of the project leader).

Four proposals have been submitted and reviewed by two external reviewers and two members of the supervisory board. The results were pretty clear, all of the proposals received good scores. However, there were two first places and two second places. Both 1st and 2nd winners were close from the score. Therefore, the overall funding budget was devided between the four groups giving all of them the chance to make this experience.

Under the header "Our goals" you find more information about this mini-proposal.

 

Our goals

The human leukocyte antigen class II (HLA-II) is a highly polymorphic protein that is mainly expressed on antigenpresenting cells (APCs) where it presents endogenous and exogenous peptides to CD4+ T cells. Different HLA-II alleles are genetically associated with inflammatory and auto-immune diseases. Nevertheless, a functional understanding of this association signal is still lacking. Recently, artificial intelligence (AI) has achieved state-of-the-art results on a wide range of problems. The rise of Mass-spectrometry-based immunopeptidomics which enables an in-depth characterization of thousands of peptides that are presented in vivo by HLA-II proteins, along with T cell characterization methods like T cell receptor repertoire sequencing (TCRSeq) and antigen-specific T cell enrichment (ARTE) has revolutionized our understanding of the immune system. However, the majority of these efforts have been directed toward characterizing tumor-associated antigens (TAA) while characterizing the microbial-derived immunopeptidomes (mIP) still lacking. Hence, our aim in this proposal is to address this problem by (i) generating mIP data from different APCs, (ii) characterizing the Tcell response towards these peptides using TCRSeq and ARTE, followed by (iii) AI-based modeling to model mIP generation and to discover rules governing the immune-response towards them.

Project leader: Hesham El Abd

Project budget: 35.000 €

Running time: April - December 2020