Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/133127
|Protein abundance prediction in bulk and single-cell transcriptomics
|Rodríguez Romera, Antonio
|Calvet Liñán, Laura
|Universitat Oberta de Catalunya (UOC)
|Omic technologies are invaluable tools to study the biological organisation of organisms. In the field of transcriptomics, scientific advancements have enabled the development of extraordinarily sensitive techniques that can measure complete transcriptomes from single cells. On the other hand, several attempts have been made to measure single cell proteomes. However, most of them lack the coverage and the sensitivity of transcriptomic techniques. Although transcriptomic tools are widely used, several studies have shown that RNA and proteins are not sufficiently correlated to act as proxies for each other. In this thesis we have explored the use of the Protein/RNA ratios to improve RNA and protein correlations. Using a bulk proteogenomic dataset we expanded the work of previous authors and showed that this ratio can be used to impute protein levels from transcriptomic abundances in several human tissues. Importantly, this strategy was independent of the tissue composition and was also applicable for cell surface proteins. Using recently published CITE-seq atlases we explored for the first time this approach in single-cell data. Our results showed that Protein/RNA ratios can better predict protein levels in single cell data when they are computed from CITE-seq datasets compared to bulk-data-calculated ratios. Interestingly, protein prediction performed well using correction factors computed from a different experiment, suggesting that this approach can be generalised to other single cell datasets.
|Appears in Collections:
|Trabajos finales de carrera, trabajos de investigación, etc.
Files in This Item:
|Memory of TFM