Towards good representations of single-cell protein expression data
Master thesis
Permanent lenke
https://hdl.handle.net/11250/3140868Utgivelsesdato
2024-06-17Metadata
Vis full innførselSamlinger
- Master theses [201]
Sammendrag
Many diseases including infections and cancer can evoke an immune response that is detectable as changes in the immune cell composition in blood.In cancers originating in the immune system, the immune cell compositioncan also change due to uncontrolled growth of the malignant cells, shiftingthe normal balance between immune cell types. One example is acute myeloid leukemia (AML), which affects a precursor of several immune cell typesfound in blood including monocytes and neutrophils. Single-cell proteinexpression measurements obtained with CyTOF offer a powerful means ofstudying immune cell composition in blood, and this thesis concerns theanalysis of such data. We specifically consider the problem of convertingsuch data - which describe which proteins are expressed on each analyzedcell - to a representation that reveals the immune cell composition. Wealso study how to obtain representations that are well suited as inputs toalgorithms for prediction of treatment outcome and survival. The focus willbe on unsupervised clustering, and we propose a novel semi-supervisedclustering algorithm and compare its performance with other methods.