Systematic evaluation with practical guidelines for single-cell and spatially resolved transcriptomics data simulation under multiple scenarios

All total 152 datasets are collected in the benchmarking study. Every dataset contains two parts: the gene expression matrix (or well-established model by dynwrap for trajectory) and the data information including the data id, repository, accession number, URL, technology platform, species, organ (source), cell number, gene number, data type, ERCC spike-in, dilution factor, volume, group condition, treatment, batch information and cluster labels. There are 23 datasets (data79-data101) for evaluating the simulation ability for cell trajectories which are derived from another Zenodo repository (https://zenodo.org/record/1443566).