Fig. 1
The overview of GepLiver workflow and main content. GepLiver curated RNA sequencing data of 2469 human bulk tissues, 492 mouse liver samples, 409,775 single cells from 347 human samples and 27 human liver cell lines in total covering 16 liver phenotypes and 2 species. RNA-seq raw reads were processed through the standardized pipeline of quality control, reads mapping and feature quantification using Assembling Splice Junctions Analysis (ASJA) and circRNA Identifier (CIRI2) algorithm whereas raw data of single cell RNA-seq were reanalyzed by CellRanger followed by downstream analysis of Seurat. Single cell datasets involved were harmonized into a liver reference map from which 16 cell types and 101 subtypes were finely identified. The expression landscape of normalized transcripts was further combined with gene dependency scores and literature metadata for functional analysis. GepLiver facilitates the visualization and direct comparison of gene expression among various liver phenotypes of bulk tissue, cell lines and single cells. The Analysis section including dependence, survival and comparison modules was provided for function explorations.
