dc.date.accessioned | 2024-01-29T13:24:56Z | |
dc.date.available | 2024-01-29T13:24:56Z | |
dc.date.issued | 2024 | |
dc.identifier.uri | http://hdl.handle.net/10852/107194 | |
dc.description.abstract | Every day, our immune system fights off potential threats such as viruses, bacteria or cancer cells. The key molecules used to recognise and memorise these antigens are so-called ‘adaptive immune receptors’. Recent technological advancements allow us to read the DNA sequences of these immune receptors, and analyse them using machine learning algorithms. Such algorithms can be used to predict the antigen binding targets of immune receptors, which may be used to design therapeutic drugs. Alternatively, one could train algorithms to diagnose a range of diseases based on a single blood sample. In this thesis, Lonneke Scheffer and colleagues have developed immuneML, a software platform for the machine learning analysis of adaptive immune receptor data. immuneML can be used to develop and compare machine learning methods for antigen binding or disease prediction. Furthermore, a tool named CompAIRR was created for the ultra-fast and memory efficient comparison of large repertoires of immune receptors. CompAIRR can be used as a stand-alone tool, but has also been used to speed up components of immuneML. Lastly, a new method was integrated into immuneML, to test the hypothesis whether antigen binding and non-binding immune receptors can be distinguished based on the presence of short motifs. | en_US |
dc.language.iso | en | en_US |
dc.relation.haspart | Paper 1. Pavlović, M., Scheffer, L., Motwani, K., et al. (2021). The immuneML ecosystem for machine learning analysis of adaptive immune receptor repertoires. Nature Machine Intelligence, 3(11), 936–944. DOI: 10.1038/s42256-021-00413-z. The article is included in the thesis. Also available at: https://doi.org/10.1038/s42256-021-00413-z | |
dc.relation.haspart | Paper 2. Rognes, T., Scheffer, L., Greiff, V., Sandve, G. K. (2021). CompAIRR: ultra-fast comparison of adaptive immune receptor repertoires by exact and approximate sequence matching. Bioinformatics, 38(17), 4230–4232. DOI: 10.1093/bioinformatics/btac505. The article is included in the thesis. Also available at: https://doi.org/10.1093/bioinformatics/btac505 | |
dc.relation.haspart | Paper 3. Scheffer, L., Reber, E. E., Mehta, B. B., Pavlović, M., Lê Quý, K., Richardson, E., Akbar, R., Greiff, V., Haff, I. H., Sandve, G. K. Predictability of antigen binding based on short motifs in the antibody CDRH3. Manuscript in preparation. To be published. The paper is not available in DUO awaiting publishing. | |
dc.relation.uri | https://doi.org/10.1038/s42256-021-00413-z | |
dc.relation.uri | https://doi.org/10.1093/bioinformatics/btac505 | |
dc.title | Machine learning and computational analyses of adaptive immune receptors | en_US |
dc.type | Doctoral thesis | en_US |
dc.creator.author | Scheffer, Lonneke | |
dc.type.document | Doktoravhandling | en_US |