Gene-expression memory-based lineage prediction of cell lineages from scRNA-seq datasets

This dataset contains the R code and source data to generate the main figures of the associated publication. For every figure panel one folder is given containing both the R code to generate the respective plot/image and the input data.

The associated publication describes the computational tool Gene-Expression Memory-based Lineage Inference (GEMLI) to annotate cells related over several cell divisions (small and mid-sized lineages) in scRNA-seq data. The basis of GEMLI is gene expression memory, the stability of expression of "memory genes" across cell divisions. Lineage assignment using GEMLI allows to study heritable gene expression, to discriminate symmetric and asymmetric cell fate decisions and to reconstruct individual multicellular structures from pooled scRNA-seq datasets. GEMLI is avialable as R package on GitHub (https://github.com/UPSUTER/GEMLI).

The code and data avalailable makes it possible to reproduce all analyses of memory genes and GEMLI lineage predictions shown in the main figures of the associated publication. This corresponds to analyses of public lineage-annotated ground truth scRNA-seq datasets of diverse cell types (mESC GSE226169, MEF GSE99915, CD8 and L1210 GSE74923, HSC GSE67317, WM989 GSE237228, pancreatic metastases GSE173958, intestinal crypts and organoids GSE140802, bone marrow cells and K562 GSE182685). Further GEMLI predictions and further analysis of one paired Xenium and scFFPE-seq dataset of human breast cancer (GSE243280) is shown. See also the associated publication.