A team led by Chinese scientists has published the first-of-its-kind three-dimensional spatiotemporal cellular atlas, which covers the entire life cycle of rice—from germinating seed to flowering and grain filling—in the journal Cell. Furthermore, the researchers provided global scientific workers with a portal and a foundational model based on single-cell rice data.
The article, titled 'A spatiotemporal lifecycle atlas decodes spatial coordination in rice,' was posted online on the evening of October 6. The corresponding institution is Southern University of Science and Technology, and Professor Hunwei Guo is listed among the authors responsible for the publication, according to data from these universities and BGI. Collaboration was carried out with the Yazhouwan National Laboratory, BGI Research, Huazhong Agricultural University, Wuhan University, and the Institute of Genetics and Developmental Biology of the Chinese Academy of Sciences, as per these universities' and BGI's reports.
The atlas was created by combining single-nucleus RNA sequencing with Stereo-seq spatial transcriptomics from BGI on the Japanese cultivar Junhua 11 and a new gapless whole-genome reference genome. It covers 10 types of organs and tissues across 61 stages. The data profiles approximately 851,725 nuclei and 347,640 spatial cells, identifying 119 cell types and 133 subtypes.
The derived developmental trajectory links cell states from early embryo to mature organs. Breeders have long faced the problem of genes that improve one trait while disrupting others. The atlas shows exactly where functional genes act in different tissues and at different stages. Partners at BGI Research developed the Rice Spatio Temporal Atlas portal for gene querying, viewing spatial expression, and comparison. The RICE scGPT model, adapted for rice, is a single-cell data-based foundational model that helps annotate cell types with similar patterns in other studies.
Among the findings highlighted by BGI, the auxin-responsive regulator OsARF1 is linked to multiple developmental branches, mostly with different predicted targets, demonstrating how broadly the impact of master gene editing can spread. Spatial maps also illustrate dorsal and ventral programs in the endosperm, which are directed either towards starch-related genes or storage protein genes as the grain fills.
China Central Television, citing IT Home, noted that this resource should promote more precise breeding design, allowing researchers to target context-specific modules instead of making crude changes. The authors present this atlas as a reusable foundation for understanding how the rice plant is built cell by cell.