Temporal Analysis of Embryonic Epidermal Morphogenesis in <i>Caenorhabditis elegans</i>

The development of the epidermis plays a central role in driving the morphogenesis of the Caenorhabditis elegans embryo. This is a four-dimensional process, tightly regulated by developmental timing. Research on epidermal morphogenesis has often emphasized the analysis of defective embryos to uncover the underlying mechanisms. However, current research may place disproportionate focus on overt phenotypic abnormalities, potentially overlooking the crucial role of developmental timing. Consequently, existing methods are limited in the early screening of genes involved in epidermal morphogenesis, as the inactivation of certain genes does not produce visible morphological defects but instead leads to subtle delays during specific embryonic stages. This highlights the need to incorporate developmental timing into conventional phenotype-based analyses. In this study, we developed a deep learning–based image analysis pipeline that combines ResU-Net for noise reduction and ResNet for embryonic stage prediction. Time-lapse differential interference contrast (DIC) microscopy was used to dynamically track the timeline of epidermal morphogenesis. Stage-specific developmental delays were observed in RNAi-treated embryos targeting ajm-1, tes-1, and leo-1, which was clearly reflected in the temporal dimension. This temporal perspective offers a novel framework for understanding the dynamics of morphogenesis and reveals subtle developmental processes.