This job is currently Archived,
Master’s Thesis: Immunogenicity Prediction - Sweden
Posted on : 03 June 2017
Master’s Thesis: In silico immunogenicity
- Our company is an international specialty healthcare company dedicated to rare diseases.
- Our mission is to develop and deliver innovative therapies and services to improve the lives of patients.
- The product portfolio is primarily focused on Haemophilia, Inflammation and Genetic diseases
- We also market a portfolio of specialty and rare disease products for partner companies across Europe, the Middle East, North Africa and Russia.
- Our company is a pioneer in biotechnology with world-class capabilities in protein biochemistry and biologic manufacturing.
- Biological drugs are not hampered by the toxic side-effects that commonly plague small molecule drugs.
- They are however at risk of evoking an immune response by detection of the immune system, and specifically by recognition of T-cells after presentation of peptides that are derived from the therapeutic protein on MHC class II receptors.
- In attempts to minimize the risks with the development of new protein therapeutics the prediction of binding to MHC class II molecules are an integral part of the development of new protein drugs.
- Currently we employ a number of separate tools in order to assess the immunogenic potential of new protein sequences.
- It would facilitate our work if we would have a tool to can perform the workflow of stepwise immunogenicity assessment an investigated protein sequence.
- Our preferred route to predict immunogenicity today is to get the probability of each 9-mer to bind to a MHC class II, score predicted high probability binders with a Janus matrix that looks for functional homology with the human genome, those hits in turn should be predicted to bind to the same HLA subtype to decrease the score of the original binder.
- Implement a web server that can be assessed internally to predict immunogenicity, currently a software called TEPredict is used for binding predictions and patmatdb for janus matrix comparisons. An approach could be to use the existing software to create a workflow for the complete analysis.
- The output should be possible to select different format for to answer different questions.
- A more extensive approach could include to use the current public binding data to create a new algorithm to predict the potential of each protein to bind to MHC class II of different sequence.
- The project will be performed in collaboration with computational biologist and immunologists at the Research & Translational Science at our company.
- Master student in Bioinformatics or Computer Science or related field
- Experience and knowledge in one or several of the following skills:
- Web server setup
- Script language
- Workflow software (KNIME or similar)