This study deciphers the gene regulatory network (GRN) underlying maize endosperm differentiation using single-cell transcriptomics. By analyzing 17,022 single cells, the researchers identified 12 cell clusters corresponding to five endosperm cell types, revealing complex transcriptional heterogeneity. They mapped the temporal gene-expression pattern from 6 to 7 days after pollination and constructed a GRN by integrating single-cell transcriptomic data with DNA-binding profiles of 161 transcription factors (TFs). This resulted in 181 regulons, containing genes encoding TFs and their high-confidence targets. The study also identified cell-cluster-specific essential regulators and validated three predicted key regulators. Maize endosperm is a critical component of cereal grains, and its development involves a series of stages, including a multinucleate coenocyte stage, cellularization, and differentiation into distinct cell types. The study provides a comprehensive transcriptome atlas and large-scale TF-DNA-binding profiles, revealing the transcriptional complexity of maize endosperm development. The researchers identified distinct cell clusters within known endosperm cell types, highlighting the heterogeneity within these compartments. The study also analyzed the function of each cell cluster using Gene Ontology (GO) annotations, revealing that different cell types and clusters within the same cell type were enriched for different GO terms. For example, the ESR cluster exhibited enrichment for "cell-cell signaling," while the AL clusters showed enrichment for "cellular respiration." The study further explored the role of phytohormones in endosperm development, showing that different cell clusters exhibited distinct patterns of phytohormone biosynthesis and response. The researchers also identified essential regulators of cell-cluster identity, including MRP1 for BETL development and O11 and NKD2 for SE development. They constructed a high-confidence GRN by integrating TF-binding profiles and coexpression-based regulatory networks, revealing 181 regulons with a median size of 70 genes per regulon. These regulons formed a scale-free network and were associated with specific biological functions, such as "peptide biosynthetic process." The study further validated the roles of key regulators in BETL development, including MYBR29, which was shown to be essential for BETL differentiation. The results highlight the complexity of endosperm development and provide a framework for understanding cereal endosperm development at single-cell resolution. The study also underscores the importance of TFs in determining cell identity and their potential as targets for crop breeding.This study deciphers the gene regulatory network (GRN) underlying maize endosperm differentiation using single-cell transcriptomics. By analyzing 17,022 single cells, the researchers identified 12 cell clusters corresponding to five endosperm cell types, revealing complex transcriptional heterogeneity. They mapped the temporal gene-expression pattern from 6 to 7 days after pollination and constructed a GRN by integrating single-cell transcriptomic data with DNA-binding profiles of 161 transcription factors (TFs). This resulted in 181 regulons, containing genes encoding TFs and their high-confidence targets. The study also identified cell-cluster-specific essential regulators and validated three predicted key regulators. Maize endosperm is a critical component of cereal grains, and its development involves a series of stages, including a multinucleate coenocyte stage, cellularization, and differentiation into distinct cell types. The study provides a comprehensive transcriptome atlas and large-scale TF-DNA-binding profiles, revealing the transcriptional complexity of maize endosperm development. The researchers identified distinct cell clusters within known endosperm cell types, highlighting the heterogeneity within these compartments. The study also analyzed the function of each cell cluster using Gene Ontology (GO) annotations, revealing that different cell types and clusters within the same cell type were enriched for different GO terms. For example, the ESR cluster exhibited enrichment for "cell-cell signaling," while the AL clusters showed enrichment for "cellular respiration." The study further explored the role of phytohormones in endosperm development, showing that different cell clusters exhibited distinct patterns of phytohormone biosynthesis and response. The researchers also identified essential regulators of cell-cluster identity, including MRP1 for BETL development and O11 and NKD2 for SE development. They constructed a high-confidence GRN by integrating TF-binding profiles and coexpression-based regulatory networks, revealing 181 regulons with a median size of 70 genes per regulon. These regulons formed a scale-free network and were associated with specific biological functions, such as "peptide biosynthetic process." The study further validated the roles of key regulators in BETL development, including MYBR29, which was shown to be essential for BETL differentiation. The results highlight the complexity of endosperm development and provide a framework for understanding cereal endosperm development at single-cell resolution. The study also underscores the importance of TFs in determining cell identity and their potential as targets for crop breeding.