Other - Data analsis; Bioinformatics
The single cell mRNA sequencing (scRNA-Seq) has become a widely used tool for immune profiling and biomarker discovery. A major bottleneck in the initial steps of scRNA-Seq data analysis is initial processing of data, which is time-consuming since it requires a lot of interaction between the end-user and the bioinformatician. We have developed a graphical user interfaced based on the Shiny/R framework (scShinyHub) that allows immunologists to intuitively and autonomously perform cell and gene selection in the process of scRNA-Seq data analysis.
The tool is open (source) and easily extendable. Oher tools with a graphical user interface mainly focus on post-differential data analysis and use all data, including cells/genes not pertinent to the biological question. scShinyHub focuses on filtering cells and genes based on various criteria and on subsequent reanalysis of the cells of interest. Another feature is the plug-in structure that allows developers to easily add functionalities.
We are presenting the scShinyHub workflow that allows to (1) remove mitochondrial/ribosomal genes and other non-interesting genes; remove cells not relevant to the biological question; (2) perform quality controls (UMI distribution, …); (3) apply dimensionality reduction (tSNE, UMAP, PCA, …) and clustering (hierarchical, self-organizing maps) methods; (4) select/remove cells based on different criteria and re-run analyses; (5) finalize the workflow by running differential expression analysis, or trajectory inference; (6) generate reports.