(computational biopharmaceutics and modeling), Pharmaceutical Chemistry, Network Pharmacology Metabolic conditions, including type 2 diabetes mellitus (T2DM), obesity, dyslipidemia, and metabolic dysfunction associated with steatotic liver disease, together with their cardiovascular and renal complications, continue to rise in prevalence and impose a substantial economic burden worldwide. Current estimates indicate that more than 500 million individuals are affected by diabetes globally and is projected to increase markedly by 2045 (Sun et al., 2022). The multi-organ and multi-pathway dysregulation characteristic of these disorders leads to clinical heterogeneity, making it difficult to select treatment targets and resulting in mixed responses.Small molecules extracted from natural products and traditional medicines remain an important source of therapeutics, particularly for pathway-level modulation and rational polypharmacology (Newman and Cragg, 2020;Atanasov et al., 2021). In parallel, systems pharmacology, multi-omics technologies, and data-driven artificial intelligence (AI) methods are transforming the field by bridging phytochemicals, molecular targets, and disease phenotypes, enabling exploration of large chemical spaces and generation of experimentally testable mechanistic hypotheses (Hopkins, 2008;Vamathevan et al., 2019).Together, these advances provide a more tractable and mechanism-based framework for natural product research, from dataset construction and model development to candidate prioritization and biological validation.The Research Topic Natural Medicines for Metabolic Diseases: Computational and Pharmacological Approaches brings together contributions that integrate computational and experimental strategies to accelerate translation. It continues the editorial vision introduced in Volume I of this series (Stalin et al., 2024). Volume II comprises six papers: three original research articles and three review or meta-analytic studies, covering diabetic complications, target discovery, and methodology development. Cautious evidence synthesis also benefits clinical translation. Lin et al. conducted a systematic review and meta-analysis of Qingre Lishi decoction (QRLSD) for T2DM, summarizing data from 18 randomized controlled trials. The analysis reports improvements in fasting and postprandial glucose, HbA1c, lipid profiles, and indices of insulin resistance, without an apparent increase in adverse events. However, substantial heterogeneity among studies and variability in formulations limit the strength of causal inference. The authors appropriately call for large, multicenter trials using standardized preparations and extended follow-up to assess generalizability and long-term benefit.In Volume II, several priorities emerge for the next stage. First, improved data infrastructure is essential: standardized chemical identifiers, harmonized endpoints, curated databases, and transparent preprocessing pipelines are prerequisites for credible AI models and reproducible systems pharmacology. Second, multi-target design should be intentional rather than incidental; multi-objective optimization, uncertainty-aware prioritization, and explicit off-target risk assessment are necessary to balance efficacy, safety, and developability. Third, the most impactful studies will tightly integrate computation and experimentation, translating omics-derived hypotheses into a small number of high-confidence targets for directionality-aware testing and subjecting AIprioritized compounds to prospective evaluation, including early ADMET screening.Finally, openness should be the norm: shared code, datasets, and validation protocols will facilitate independent replication and rapid methodological iteration.Together, the six articles create a cohesive impact by supporting a realistic message: finding natural medicines to treat metabolic disorders is most effective when computational prioritization, inference of systems-level mechanisms, and pharmacological validation are integrated into a single auditable process. The Volume II collection is expected to help increase reproducible, mechanism-based pipelines and accelerate the development of promising natural products into plausible therapeutic candidates.
Stalin et al. (Fri,) studied this question.