Browsing by Subject "Shiny app"

Now showing 1 - 1 of 1

Embargo
Development of a multi-species scRNA-seq atlas of metabolic dysfunction–associated steatotic liver disease (MASLD)
(2024-07) Demirdizen, Mert
Metabolic dysfunction-associated steatotic liver disease (MASLD) affects a significant portion of the human population, potentially leading to severe secondary conditions. Despite extensive research on patients and animal models to understand the disease's initiation and progression, and the development of various therapeutic strategies, current non-invasive techniques remain inefficient, and only one drug has been approved. This underscores the critical need to elucidate the cellular mechanisms driving MASLD to develop new diagnostic and therapeutic approaches. Single-cell RNA sequencing (scRNA-seq) enables the investigation of transcriptomic profiles at the cellular level. Various studies have utilized scRNA-seq to explore MASLD in patient samples and animal models. Although there are numerous publicly available scRNA-seq datasets from these studies, no comprehensive tool exists to explore and analyse them collectively. In this study, publicly available scRNA-seq datasets related to MASLD were collected and reanalysed, adhering to best practices suggested in the literature, and a web tool was developed for their visualisation and custom downstream analyses. Thirteen datasets obtained from the livers of human MASLD spectrum patients, and three datasets different animal models, two mouse and one zebrafish, were uniformly preprocessed. The two human datasets containing cirrhotic samples and two fibrotic datasets, one from mouse and another from zebrafish models, were integrated to reveal the conserved mechanisms of liver fibrosis among human patients and animal models. A web tool was developed in R Shiny to enable visualizations, cell cluster selection, differential expression analysis and gene set enrichment of genes and gene sets. The utility of the app was demonstrated by conducting a case study with the integrated dataset. The results included an endothelial cluster that had cells from each of the four datasets, and differential gene expression analysis on the integrated dataset provided significantly modulated known, e.g., SYNPO and TIMD4 liver fibrosis-associated genes across the three species. As a result, the custom analysis workflow provided in this thesis can allow the detection of conserved markers in different MASDL stages by further analysing this and other integrated datasets in the future.