Modeling human diseases using the fruit fly (Drosophila melanogaster) has established itself as a cornerstone of functional genomics and preclinical medicine. Despite its anatomical simplicity, the Drosophila genome shares remarkable functional conservation with human disease-related genes, enabling the study of complex physiological traits through accessible tissue models. Furthermore, beyond individual disease models, we propose a framework demonstrating how these diseases converge at common molecular centers, such as the breakdown of protein homeostasis, mitochondrial dysfunction, chronic inflammation, and organ-to-organ communication. Finally, we discuss strategies for integrating the Drosophila platform into drug development pipelines and establishing standards to enhance inter-laboratory reproducibility. Overall, this review highlights the enduring value of fruit flies as a model system, particularly when combined with AI-omics approaches to transform complex biological datasets into actionable therapeutic strategies.
The standardized construction of ethograms and the quantitative comparison of behavioral differences are fundamental to animal behavioral ecology research. This study focused on the Yunnan snub-nosed monkey (Rhinopithecus bieti) inhabiting the Weixi area of the Baimaxueshan National Nature Reserve and the Mangkang area of the Mangkang Yunnan Snub-nosed Monkey National Nature Reserve. From 2017 to 2021, a total of 120 infrared cameras were deployed, yielding 21,813 image and video files containing R. bieti. Based on the PAE (Posture–Act–Environment) coding system, an ethogram and corresponding coding framework were constructed. In total, 16 postures, 84 acts, and 22 environmental factors were identified; 135 behaviors were defined and classified into 18 functional categories. On this basis, behavioral composition differences were quantitatively compared across seasons and geographic regions. The results showed that season significantly affected behavioral composition (F = 5.029, R2 = 0.255, p < 0.01). Four behavioral categories exhibited significant seasonal fluctuations: Locomotive behavior (χ2 = 12.97, p < 0.01), Ingestive behavior (χ2 = 12.04, p < 0.01), Resting behavior (χ2 = 26.18, p < 0.001), and Miscellaneous behavior (χ2 = 29.92, p < 0.001). Locomotive behavior peaked in summer, whereas Ingestive behavior was at its lowest and Resting behavior at its highest in autumn. Geographic region also significantly influenced behavioral composition (F = 2.808, R2 = 0.149, p < 0.01). These differences were mainly reflected in significantly lower Locomotive behavior (U = 13, p < 0.01) and higher Ingestive behavior (U = 13, p < 0.05) in the higher-elevation Mangkang group compared with the lower-elevation Weixi group. By integrating infrared camera with the PAE coding system, this study not only constructs an ethogram for R. bieti but also compares behavioral differences across seasonal and geographic dimensions. The findings demonstrate the comprehensive applicability of this technical pathway for macroscopic behavior recognition, frequency statistics, and difference analysis in wild primates, and provide fundamental data for further understanding the ecological driving mechanisms of behavioral plasticity and for differentiated conservation management of R. bieti.
Complex diseases, including cancer, rare genetic disorders, neurodevelopmental and psychiatric conditions, and neurodegenerative diseases, arise from interactions among genetic variation, gene regulation, and cellular states that are difficult to capture using a single data type or biological scale. Biological foundation models address this challenge by treating nucleotides and genes as tokens and learning representations that can be transferred to downstream biomedical and clinical tasks. In this review, we examine two major model classes, genomic sequence foundation models and cell foundation models, and compare their tokenization strategies, model architectures, pretraining objectives, and adaptation methods. We summarize their emerging applications in regulatory variant interpretation, disease-associated cell-state analysis, drug-response prediction, and therapeutic target discovery across complex diseases. We distinguish applications supported by experimental or retrospective validation from those that remain primarily computational or conceptual. We further discuss key challenges to clinical translation, including multimodal data integration, model interpretability, benchmarking, patient-specific prediction, and privacy protection. We highlight future opportunities to integrate biological foundation models with emerging frameworks of medical digital twins, agentic AI, and federated learning. By linking model design to translational goals, this review provides a practical framework for evaluating biological foundation models and their readiness for complex disease research and clinical use.
The megaspore mother cell (MMC) is the founding cell of the plant female germline, and its specification is a critical step in sexual reproduction of flowering plants. Recent advances in single-cell transcriptomics, live imaging, and gene editing have significantly advanced our understanding of the molecular mechanisms underlying MMC specification in Arabidopsis thaliana. Here, we systematically review the latest findings from three integrated perspectives. Cell-autonomous regulatory mechanisms that establish MMC fate; non-cell-autonomous signals originating from somatic cells that position and permit MMC specification; and mechanisms by which neighboring somatic cells maintain their own fate and restrict MMC number. We also outline key open questions and offer perspectives on future research directions. This review provides a comprehensive reference for understanding the molecular-level mechanisms governing female germ cell fate determination in plants.
Smut disease is a major constraint on proso millet (Panicum miliaceum L.) production. Our previous metabolomic work in the cultivar ‘Chishu 13’ outlined a two-stage defense strategy involving energy metabolism at jointing and phenylpropanoid activation at heading, yet the transcriptional regulation underlying this metabolic shift remained unknown. In this follow-up study, we performed RNA-seq on the same biological samples used in our prior metabolomics analysis to directly link transcriptomic changes to metabolic outcomes. We compared asymptomatic inoculated (IA) and symptomatic inoculated (IS) plants across four developmental stages (seedling, tillering, jointing, and heading) using time-series profiling, weighted gene co-expression network analysis (WGCNA), and multi-omics integration. Transcriptional reprogramming was minimal during early infection but escalated sharply from jointing (5498 differentially expressed genes) to heading (6818 DEGs), matching the previously observed metabolic divergence. K-means clustering revealed two distinct transcriptional programs: TCA cycle and oxidative phosphorylation pathways dominated at jointing, while phenylpropanoid biosynthesis and plant–pathogen interaction pathways peaked at heading. By combining WGCNA with temporal clustering, we identified 337 high-confidence Stage 1 and 408 Stage 2 candidate genes. Seven hub genes encoding TCA cycle enzymes correlated positively with citrate and succinate accumulation only in IA plants at jointing, whereas eight phenylpropanoid-related hub genes strongly associated with L-phenylalanine and cinnamic acid levels in IA plants at heading. These coordinated transcriptional–metabolic modules were disrupted in IS plants, and RT-qPCR confirmed phenotypic-group-specific expression of key hub genes. This study presents a temporally resolved transcriptional atlas of proso millet response to Anthracocystis destruens, extending beyond prior single-time-point multi-omics studies by integrating matched transcriptomic and metabolomic time-series data to establish a dynamic multi-omics framework for smut resistance. We demonstrate that effective defense requires precise sequential coordination of energy and phenylpropanoid metabolism, providing validated hub genes as targets for molecular breeding.
