A unified framework for realistic in silico data generation and statistical model inference in singl

Dongyuan Song, PhD candidate in Bioinformatics, UCLA Qingyang Wang, PhD student in Statistics, UCLA Guanao Yan, PhD candidate in Statistics, UCLA Tianyang Liu, Master of Applied Statistics, UCLA Jingyi Jessica Li, Associate Professor in the Department of Statistics, Department of Human Genetics, Department of Computational Medicine and Department of Biostatistics, UCLA

In the single-cell and spatial omics field, computational challenges include method benchmarking, data interpretation, and in silico data generation. To address these challenges, we propose an all-in-one statistical simulator scDesign3 to generate realistic single-cell and spatial omics data, including various cell states, experimental designs, and feature modalities, by learning interpretable parameters from real datasets. Furthermore, using a unified probabilistic model for single-cell and spatial omics data, scDesgin3 infers biologically meaningful parameters, assesses the quality of cell clusters and trajectories, and generates in silico controls for benchmarking computational tools.