Dissecting cell identity via network inference and in silico gene perturbation

Cell identity is governed by the complex regulation of gene expression, represented as gene-regulatory networks(1). Here we use gene-regulatory networks inferred from single-cell multi-omics data to perform in silico transcription factor perturbations, simulating the consequent changes in cell identity using only unperturbed wild-type data. We apply this machine-learning-based approach, CellOracle, to well-established paradigms-mouse and human haematopoiesis, and zebrafish embryogenesis-and we correctly model reported changes in phenotype that occur as a result of transcription factor perturbation. Through systematic in silico transcription factor perturbation in the developing zebrafish, we simulate and experimentally validate a previously unreported phenotype that results from the loss of noto, an established notochord regulator. Furthermore, we identify an axial mesoderm regulator, lhx1a. Together, these results show that CellOracle can be used to analyse the regulation of cell identity by transcription factors, and can provide mechanistic insights into development and differentiation.

Automated summary

Cell identity is controlled by the regulation of gene expression in complex gene-regulatory networks. A machine-learning-based approach, CellOracle, is used to simulate transcription factor perturbations and predict changes in cell identity based on single-cell multi-omics data. This approach successfully models reported changes in phenotype caused by transcription factor perturbation and identifies previously unreported phenotypes in zebrafish embryogenesis. The results demonstrate that CellOracle can analyze the regulation of cell identity by transcription factors and provide insights into development and differentiation.