Artificial Intelligence (AI) is increasingly being applied in disease diagnosis and has shown considerable potential for improving diagnostic accuracy, efficiency, and accessibility. This narrative review aims to examine recent advances in AI-assisted disease diagnosis from the perspective of the clinical diagnostic pathway, with particular attention to how AI is applied across successive stages of disease diagnosis and clinical practice. It synthesizes recent advances in AI-assisted disease diagnosis from representative studies. This review is structured around the diagnostic pathway, encompassing disease detection and classification, risk stratification and prediction, and finally, clinical integration. The article critically examines the role of AI in three core areas: (1) the application of multimodal data for the detection and classification of non-communicable diseases, including neurological disorders and cancer; (2) genetic risk stratification and diagnosis of Mendelian disorders; (3) advancements in medical image analysis through multimodal fusion, radiomics, and real-time assistance. A central theme of this review is that AI is evolving from a single-modal detection tool toward broader clinical applications, with current evidence supporting its role primarily as a decision-support technology rather than as a replacement for clinical judgment. However, this transition is constrained by significant challenges, including data quality, model generalizability, clinical validation, and ethical and regulatory hurdles. By synthesizing these advances and limitations, this review aims to provide clinicians and researchers with a balanced perspective, thereby offering insights into the development of a trustworthy, next-generation diagnostic ecosystem.
Mesenchymal stem cells (MSCs) have attracted considerable attention for clinical translation in regenerative medicine, primarily due to their validated paracrine effects, prominent immunomodulatory properties, and superior multipotent differentiation capabilities. However, the limited homing efficiency and poor post-transplant survival of MSCs severely compromise therapeutic efficacy, thereby giving rise to suboptimal and inconsistent treatment outcomes. To circumvent these critical drawbacks and fully harness the therapeutic potential of MSCs, researchers have incorporated a diverse array of effective strategies, including genetic engineering, preconditioning with cytokines, small molecular compounds or hypoxic stimuli, and scaffold-based culture systems. Given these promising research advances, this review systematically summarizes recent advances in MSC-enhanced therapeutic strategies, and elaborates on their core molecular mechanisms as well as how these mechanisms modulate MSC survival, homing capacity and immunomodulatory efficacy. On this basis, we further analyze the practical applicability of these enhanced MSCs in clinical trials and seek to provide critical insights for the clinical selection of MSC-based enhanced therapies.
Microbial lipases are versatile biocatalysts with high catalytic efficiency, substrate specificity, stability, and ability to catalyze a wide range of processes under mild environmental conditions, which make them highly valuable in various industrial and biotechnological applications. However, traditional methods of enzyme discovery and engineering rely on cultured microorganisms and labor-intensive experimental processes. This study highlights recent developments in metagenomics and AI technologies for microbial lipase discovery and engineering and providing a brief overview of the sources, structural features, physicochemical properties, and industrial applications of lipases. Recent breakthroughs in metagenomics have provided new access to novel enzymes from non-cultivable microbial communities, and the rising significance of artificial intelligence in enzyme discovery, structure prediction, protein engineering, and bioprocess optimization is presented. This study also highlights the important synergy between metagenomics and artificial intelligence technologies for the identification and rational design of enzymes, integrating extensive sequence databases with predictive computational modeling tools. In addition, there are still various challenges, such as low heterologous expression levels, a lack of quality information, and limited industrial-scale validation. We anticipate that future advances in protein language models, generative artificial intelligence, synthetic biology, and multi-omics integration will accelerate enzyme discovery, engineering, and large-scale industrial implementation. Overall, the use of metagenomics, artificial intelligence, and experimental approaches has tremendous potential for developing efficient and economically viable lipases for sustainable biotechnological applications.
Cocoa butter (CB) is the defining lipid of chocolate, valued for its unique triacylglycerol (TAG) composition and polymorphic behaviour that confer gloss, snap, and rapid melt-in-mouth properties. Increasing volatility in cocoa (Theobroma cacao) supply, driven by climate pressures and rising demand, has intensified interest in cocoa butter substitutes (CBS). This review summarizes recent advances across the main technological routes for CBS development. Conventional sources such as palm mid-fractions and shea butter provide compositional compatibility, while enzymatic interesterification and fractionation can partially reshape TAG profiles towards CB-like structures. Engineered oleaginous yeasts have also been developed to produce CBS lipids through metabolic rewiring and heterologous acyltransferase expression, although economic constraints remain. Plant metabolic engineering has generated seed oils enriched in oleic, stearic, or palmitic acids with reduced polyunsaturated fatty acids. However, reproducing the balanced palmitic-stearic-oleic composition and stereospecific TAG architecture of CB in scalable crop systems remains a major challenge.
Nanoplastics (< 1 μm) represent a pervasive class of environmental contaminants with unique physicochemical properties that profoundly influence microbial ecosystems. Their high surface-area-to-volume ratio, weathering-induced functionalization, and ability to adsorb chemical pollutants and biomolecules facilitate intricate interactions with bacterial communities. This review systematically examines nanoplastic-bacteria interactions, highlighting mechanisms such as oxidative stress induction, membrane perturbation, DNA damage, metabolic reprogramming, biofilm modulation, and enhanced horizontal gene transfer, which collectively reshape microbial structure and function. Emphasis is placed on the plastisphere microbiome as a dynamic hotspot for pollutant accumulation, pathogen enrichment, and resistance gene exchange. Bacterial biodegradation pathways, including enzymatic hydrolysis, oxidative processes, biosurfactant-mediated interactions, and multispecies consortia activity, are analyzed in detail. Advanced analytical tools, such as nanoscale imaging, spectroscopy, flow cytometry, meta-omics, and AI-assisted computational modeling, are discussed for their role in elucidating nanoplastic-microbe dynamics. Environmental and human health implications, including microbiome disruption, immunotoxicity, and ecological perturbations, are evaluated. Finally, emerging biotechnological strategies for enhancing biodegradation are explored, and critical research gaps are identified. This review provides a comprehensive framework for understanding nanoplastic-bacteria interactions, offering strategic insights for environmental monitoring, risk assessment, and bioremediation development.
Methanogenic archaea (methanogens) are key drivers of global carbon cycling and biomethane production, thriving in diverse anaerobic environments through specialized metabolic and transport systems. While methanogenesis is well understood, transport mechanisms underlying nutrient uptake, ion homeostasis, and macromolecule translocation remain poorly understood. This review compiles current knowledge on transporter proteins and substrate uptake in methanogens. It also highlights fundamental knowledge gaps and outlines experimental procedures for identifying new transporters experimentally and bioinformatic strategies to identify transporter-encoding genes. These approaches enable the investigation of cultured and uncultured methanogens, providing a broader view of transporter diversity and global distribution. Advancing transporter research will enhance insights into archaeal physiology and supports further developing biotechnological applications of methanogens for biofuels and chemical production, and sustainable energy systems.
ε-Poly-l-lysine (ε-PL) is transitioning from a food preservative to a bio-based cationic functional scaffold with opportunities spanning food, agriculture, and biomaterials. However, progress in the field is hampered by a persistent disconnect between titer-driven manufacturing optimization and performance-driven application development. To bridge this gap, we propose a specification-driven (spec-driven) framework that uses critical quality attributes (CQAs) to link molecular structure, industrial production, and application translation. First, we define a minimal, actionable set of ε-PL CQAs and map their molecular determinants to functional relevance. Next, we delineate the biological feasibility windows and constraints that govern these CQAs under high-throughput biosynthesis. We then synthesize integrated strain, process, and recovery engineering strategies to illustrate how CQAs can be translated into reproducible industrial specifications. Finally, application requirements are reverse-mapped to the CQA combinations most frequently required across use scenarios. We conclude that while titer remains important, further gains in titer alone are insufficient to unlock broad translation. Among the proposed CQAs, chain-length distribution, impurity profile, and manufacturing consistency are particularly critical because they determine functional performance, safety boundaries, and lot-to-lot reproducibility. Accordingly, chain-length control, impurity reduction, and integrated strain-process-downstream control represent key rate-limiting steps for developing specification-grade ε-PL product families.
Tuberculosis (TB), caused by Mycobacterium tuberculosis (Mtb), remains a major global health challenge due to persistent diagnostic gaps. CRISPR-Cas-based diagnostics have emerged as highly sensitive and programmable platforms for nucleic acid detection, enabling rapid identification of Mtb targets, including drug-resistance-associated mutations. These systems integrate isothermal amplification, diverse Cas effectors, and multiple signal readout strategies to achieve high analytical performance. This review provides a comparative analysis of clinically evaluated CRISPR-based TB diagnostic platforms, highlighting substantial variability in assay design, performance, and translational readiness. While many platforms demonstrate strong analytical sensitivity, their implementation remains constrained by workflow complexity and limited integration into true point-of-care formats. This highlights that successful clinical translation of CRISPR-based TB diagnostics is determined more by real-world adaptability than by analytical performance alone. The current review presents a comparative analysis of CRISPR-based diagnostic platforms for tuberculosis, evaluating the variability in assay design, analytical and clinical performance, and translational readiness across currently available systems.
Parkinson's disease is a neurodegenerative condition characterized by the accumulation of misfolded and aggregated α-synuclein in Lewy bodies and neurites. These protein aggregates contribute to neurodegeneration and motor symptoms such as bradykinesia, rigidity, and tremor. While the ubiquitin-proteasome system degrades soluble α-synuclein, aggregated and oligomeric forms are primarily cleared via the autophagy-lysosomal pathway. Mutations of the SNCA gene exacerbate α-synuclein aggregation and significantly impair its clearance, highlighting the importance of targeting toxic α-synuclein species. Strategies such as promoting autophagosome formation via 5'-AMP-activated protein kinase (AMPK) and mechanistic target of rapamycin complex 1 (mTORC1) or facilitating autophagosome maturation via RAB7-a member of the RAS oncogene family-and related effectors, have shown promise in enhancing autophagy and reducing α-synuclein pathology. Pharmacological agents such as rapamycin, trehalose, and nilotinib have demonstrated preclinical efficacy in enhancing α-synuclein clearance and alleviating disease features. Concurrently, immunotherapy approaches, including passive and active immunization, aim to enhance the immune system's ability to recognize and eliminate toxic α-synuclein species. Emerging strategies such as peptide-based therapies aim to inhibit aggregation or promote degradation of α-synuclein. At the same time, nanotechnology enables the targeted delivery of therapeutic agents across the blood-brain barrier with improved efficiency. Additionally, novel AUTOTAC (autophagy-targeting chimera) platforms offer a precision strategy to tag α-synuclein for autophagic degradation. This review explores many advances in autophagy-mediated aggregated α-synuclein clearance, emphasizing its potential as a therapeutic strategy to address the limitations of current symptomatic treatments and slow the progression of Parkinson's disease.
Surface-layer (S-layer) proteins, forming the outermost envelope of many bacteria and archaea, exhibit extraordinary structural precision and self-assemble into two-dimensional crystalline lattices with square, hexagonal, or oblique symmetry. These monomolecular arrays, typically 5 to 25 nm in periodicity (varying by species), offer defined porosity and serve as robust biological nanoplatforms. Their innate capacity for self-assembly and molecular ordering has attracted significant attention in nanobiotechnology, vaccine development, biosensing, drug delivery, and ultrafiltration. S-layers are especially valued for their ability to mimic viral capsids, enhance antigen presentation, stabilize lipid bilayers, and provide highly organized scaffolds for enzyme immobilization and nanopatterning. Recent experimental achievements include the use of S-layer fusion proteins for mucosal vaccine delivery and the development of recombinant S-layer-based electrochemical biosensors. However, transitioning these advances to commercial-scale applications remains challenging. Limitations include the scalability of high-purity protein production, cost-effective recombinant expression, stability under harsh industrial conditions, and unresolved regulatory pathways for biologically derived nanomaterials. Additionally, synthetic alternatives present practical and economic competition. Nonetheless, interdisciplinary efforts in synthetic biology, materials science, and computational modeling are addressing these bottlenecks. Innovations such as cross-linkable domains, fusion with polymers or lipids, and predictive structure-function modeling are improving the robustness and adaptability of S-layer systems. As current research advances from theoretical potential to functional prototypes, S-layer proteins offer transformative prospects across medical, industrial, and environmental domains. This review uniquely integrates mechanistic S-layer biology with engineering-for-manufacture, protein-design workflows, and commercialization roadmaps - offering actionable protocols and benchmarks not covered in prior syntheses.
Optogenetics, a biotechnology that combines optical and genetic strategies to regulate cellular and tissue functions with high spatial and temporal precision, serves as a powerful tool-level regulatory technology widely employed to investigate cellular processes and elucidate disease mechanisms. Nanotechnology-driven optogenetics, which incorporates nanomaterials/nanostructures to improve the efficacy and broaden the applications of optogenetic systems, synergistically integrates the precision, tunability, and multifunctionality of nanotechnology with the spatiotemporal control inherent to optogenetics. The development not only enhances the flexibility and efficiency of optogenetic technology but also advances the field toward therapeutic-grade interventions. In this review, we summarize recent advances in nanotechnology-driven optogenetics, with a particular emphasis on three key areas: (1) nanostructured light sources for the precise activation of optogenetic systems, which include both externally light-stimulated systems and systems that operate independently of external light sources; (2) nanotechnology-enabled targeted delivery of light-sensitive proteins and genetic constructs to ensure efficient modulation of optogenetic pathways, which specifically involves the nanotechnology-assisted gene, protein, and recombinase enzyme delivery approaches; and (3) nanotechnology-driven therapeutic applications of optogenetics, including CAR T cell immunotherapy, cancer treatment, neurological interventions, and cardiac therapies. We further discuss the current challenges facing this emerging field and outline future research directions. This review aims not only to highlight recent breakthroughs but also to position nanotechnology-driven optogenetics as a promising tool for next-generation precision medicine.
Sex determination in Cannabis sativa L. has major implications for breeding, biotechnology, and crop management. In this species, sex is primarily governed by an XX/XY sex chromosome system, with genetic regulation constituting the principal level of control. Variation in sex expression, encompassing dioecious, monoecious, or hermaphroditic forms, may arise from interactions between sex chromosomes and autosomal factors X-to-autosome (X: A) chromosome balance. Secondary modulation of sex phenotype involves environmental and physiological influences, with phytohormonal regulation representing an important layer of sex plasticity. Epigenetic mechanisms may further contribute to the downstream fine-tuning of phenotypic sex expression. Importantly, sex expression in Cannabis can also be manipulated chemically. Several silver-based compounds, such as silver thiosulfate, are widely used to induce staminate flower formation in genetically female plants, whereas application of ethephon promotes female flower formation in genetically male individuals. Although both agents act through ethylene signalling pathway, their effects likely involve broader and more complex interactions within the molecular network, including cross-talk with other hormonal pathways. Within this context, transcriptional regulators associated with auxin signalling, including Aux/IAA - TOPLESS-RELATED co-repressors such as TPR1, may contribute to the activation of male developmental programmes by modulating hormonal response pathways. However, their precise role and position within the sex-regulatory hierarchy are yet to be clarified. The multi-layered regulation of sex determination in Cannabis sativa L. provides a valuable framework for investigating the molecular basis of this process through the integration of genetic, environmental, and hormonal factors. A more comprehensive understanding of these interacting regulatory layers may facilitate the development of strategies for early sex identification, improved control over flower sex conversion, and the optimisation of cannabinoid-rich female inflorescence production. Such advances would support both fundamental and applied approaches in Cannabis biotechnology and cultivation.
Cyanobacteria are increasingly positioned as photosynthetically powered, genetically tractable chassis for next-generation environmental remediation-operating as living filters that couple solar energy capture to active detoxification and resource recovery. This review synthesizes current advances in cyanobacteria-based remediation of heavy metals, micro- and nanoplastics, pathogens, and persistent organic pollutants, with particular emphasis on metabolic mechanisms, bioengineering strategies, and practical environmental applications. This review outlines a bioengineering roadmap for deploying cyanobacteria in wastewater and impacted aquatic systems to sequester and reclaim toxic heavy metals, trap nano/microplastics, attenuate pathogenic microorganisms, and chemically degrade recalcitrant organic pollutants. In the field of metal capture, recent advances in "living materials" have enabled the embedding of cyanobacteria in regenerable matrices for efficient removal and subsequent reclamation. Mechanistic insights into species such as Synechocystis have clarified adsorption behavior and stress-response determinants for cadmium and related metals, defining tunable targets including transporters, exporters, and chelation modules for strain improvement. Cobalt and uranium handling can now be rationally engineered by rewiring metal homeostasis systems or exploiting high-capacity biosorption using scalable biomass platforms like Spirulina. Beyond metals, cyanobacterial extracellular polymeric substances (EPS) are being leveraged as engineered bio-based flocculants to remove polystyrene micro- and nanoplastics, while consortia-based designs are emerging to facilitate polymer transformation. Collectively, these advances motivate the development of modular, field-ready cyanobacterial platforms immobilized, sensor-guided, and biocontained that integrate pollutant capture and circular recovery within sustainable photobioremediation pipelines. However, significant challenges remain, including field-scale validation, environmental variability, biosafety considerations, biomass management, economic feasibility, and regulatory constraints. Addressing these limitations will be essential for the practical implementation of cyanobacterial remediation technologies.
Oral and periodontal diseases are among the most prevalent chronic disorders worldwide and arise from complex interactions between microbial biofilms and dysregulated host immune responses. Despite significant advances in conventional therapies, clinical management remains challenging because of limited drug penetration, rapid clearance within the oral cavity, and poor patient compliance. In this context, microneedle (MN)-based drug delivery systems have emerged as promising minimally invasive platforms for localized and controlled therapeutic delivery. However, the adaptation of MN technologies from transdermal to oral applications requires application-specific redesign owing to the unique anatomical, physiological, and mechanical characteristics of oral tissues. Current literature also remains fragmented, particularly regarding the integration of oral tissue biology, advanced manufacturing strategies, and multifunctional MN design. This review provides a comprehensive and critical analysis of MN systems for oral and periodontal applications, with particular focus on the enabling role of additive manufacturing (AM). First, the biological characteristics of oral tissues and their implications for drug delivery are discussed, followed by an overview of MN technologies, biomaterials, and fabrication approaches. Particular emphasis is placed on oral application-specific design considerations, including mechanical constraints, penetration depth, bioadhesion, retention, and controlled drug release. Emerging therapeutic applications, ranging from antibacterial and anti-inflammatory therapies to immunomodulatory and regenerative strategies, are also critically evaluated. Recent advances in 3D-printed MNs are highlighted, emphasizing their potential for customizable architectures, integrated drug delivery systems, and patient-specific therapeutic platforms. In parallel, major translational challenges, including mechanical reliability, retention under salivary conditions, regulatory complexity, and manufacturing scalability, are critically discussed. Future perspectives involving the integration of artificial intelligence (AI), smart biomaterials, biosensor technologies, and precision medicine approaches are also explored. Beyond summarizing recent advances, this review identifies the key scientific challenges, current knowledge gaps, and emerging engineering strategies required for the successful clinical translation of 3D-printed oral and periodontal MN systems. Overall, it provides a critical roadmap for the rational design and development of next-generation personalized MN platforms, highlighting the convergence of advanced biomaterials, biofabrication technologies, and precision medicine as a foundation for future oral healthcare.
Accurate chromosome segregation in eukaryotic cells ensures that each daughter cell receives the same chromosome complement during cell division. Key to this process is the kinetochore, a macromolecular complex that connects spindle microtubules to the chromosome. The kinetochore is composed of a diverse array of proteins that can be broadly separated into an inner and outer kinetochore. Each chromosome also contains a distinct site that specifies the location of kinetochore formation, the centromere, which in many species is epigenetically defined by the specialized histone H3 variant centromere protein A (CENP-A). In this Review, we discuss recent advances in determining the complete structure and function of the inner kinetochore, as well as the sequence and 2D and 3D organization of centromeric chromatin. We discuss the implications of these recent advances for our understanding of the assembly and function of the inner kinetochore. We refer readers to a complementary review published in the same issue for a discussion on the structure and function of the outer kinetochore. Finally, we outline key questions persisting in the field driving future studies of the kinetochore and centromere.
Structural biology is undergoing a transformative era driven by advances in artificial intelligence (AI)-based protein structure prediction and cryo-electron microscopy. Predictive approaches have dramatically expanded structural coverage across proteomes and are increasingly integrated into experimental workflows. However, protein function frequently depends on dynamic molecular processes including ligand-dependent conformational remodeling, transient interactions, cooperative assembly, and transport-state transitions that remain difficult to define from static computational models alone. These challenges are particularly evident in plants, where signaling pathways often involve environmentally responsive receptor complexes, lineage-expanded regulatory proteins, and transient assemblies. Here, we discuss how experimental structural biology continues to advance plant biology by revealing mechanisms underlying hormone perception, immune receptor activation, transporter function, and enzymatic regulation. From early landmark discoveries such as the crystallization of urease to recent cryo-electron microscopy studies of dynamic signaling complexes, plant systems have repeatedly uncovered molecular architectures, chemically modified intermediates, and regulatory principles that require direct structural and biochemical characterization. Plant proteins also remain markedly underrepresented in structural databases, leaving many plant-specific pathways structurally unresolved. Together, these observations highlight the continuing importance of experimental structural biology for defining biologically relevant molecular states and enabling structure-guided strategies for crop improvement and agricultural biotechnology.
Transaminases (TAs) represent a class of enzymes that enable highly stereoselective amination reactions under mild conditions, providing an exceptionally powerful and versatile green chemistry tool for the synthesis of chiral amines and nitrogen-containing compounds. However, most current reviews on transaminases focus on the efficient synthesis of chiral amines, whereas the efficient synthesis of complex natural products has rarely been reported. This review systematically discusses the characteristics of different types of transaminases, covering their underlying catalytic mechanisms, the dynamics and roles of the cofactor involved, and key enzymatic properties that are critical for their function and practical applicability. Recent advances in protein engineering strategies and in the design of multi-enzyme cascade systems are also discussed. A particular focus is directed toward representative applications of transaminase-mediated multi-enzyme cascade systems in natural product synthesis. Based on a systematic review of recent advances in transaminase research, this review focuses on their representative applications in multi‑enzyme cascade systems for natural product synthesis, providing a comprehensive and referenceable framework for green synthetic strategies.
The publication of the Saccharomyces cerevisiae genome sequence thirty years ago marked a defining shift in modern biology, establishing yeast as the first fully sequenced eukaryotic model cell. Access to a complete reference genome has catalysed a renaissance in experimental biology, providing the foundation for advances in functional genomics, systems biology, synthetic biology, and biotechnology. This article celebrates the achievements of the yeast sequencing project and highlights how genomics has evolved from a descriptive resource into a central enabling infrastructure for engineering biology, leading to a post-genomic era defined by genome-scale design. We chart the progression from decoding the genome to the omics revolution, highlighting both recent discoveries and persistent enigmas surrounding the 'dark matter' of the yeast genome. We then describe how access to a well-annotated reference genome has culminated in a new era of synthetic genomics exemplified by the Sc2.0 project. Finally, we explore how emerging trends in artificial intelligence and bioelectronics may shape the next thirty years of yeast genomics and engineering biology. This integration of past achievements and future trajectories reinforces the enduring role of S. cerevisiae as a primary eukaryotic model for biological discovery and biotechnological innovation.
Lignin is one of the most abundant yet recalcitrant biopolymers in nature, and its inefficient depolymerization remains a major bottleneck in lignocellulosic biomass valorization and sustainable biorefinery development. Although fungal oxidative enzymes have emerged as promising biocatalysts for lignin degradation, current knowledge remains fragmented across fungal diversity, catalytic mechanisms, enzyme engineering, and industrial translation. This review critically integrates recent advances in fungal ligninolytic systems, focusing on the structure, catalytic mechanisms, and synergistic interactions of major oxidative enzymes, including lignin peroxidase, manganese peroxidase, versatile peroxidase, laccase, and auxiliary oxidases such as aryl alcohol oxidase and glyoxal oxidase. Particular emphasis is placed on the comparative catalytic efficiency, operational limitations, and industrial feasibility of these enzymes in lignin depolymerization and biomass conversion. Recent developments in protein engineering, directed evolution, heterologous expression, immobilized enzyme systems, synthetic biology, and AI-assisted enzyme optimization are also discussed as emerging strategies to improve enzyme stability, substrate specificity, and process scalability. The review further highlights the role of fungal oxidative enzymes in bioremediation, lignocellulosic biorefineries, wastewater detoxification, biofuel production, and synthesis of value-added aromatic compounds. Importantly, it establishes a direct link between fungal ligninolytic systems and circular bioeconomy models through sustainable waste valorization and biomass-to-value conversion. Overall, this review provides an integrated perspective on the challenges, technological advancements, and future prospects of fungal oxidative enzymes for sustainable industrial biotechnology.
Lung cancer remains a leading cause of cancer-related mortality worldwide due to its extensive molecular heterogeneity, late-stage diagnosis, and therapeutic resistance. Advances in high-throughput omics technologies have enabled comprehensive characterization of tumors across multiple biological layers, including genomics, epigenomics, transcriptomics, proteomics, and metabolomics. However, single-omics analyses provide only fragmented insights into tumor biology, highlighting the need for integrative multiomics approaches. Artificial intelligence (AI), particularly machine learning and deep learning, has emerged as a powerful tool for integrating heterogeneous datasets and uncovering biologically and clinically relevant patterns. This review summarizes recent advances in AI-driven multiomics integration for lung cancer, highlighting its applications in molecular subtyping, biomarker discovery, prognosis prediction, therapeutic response modeling, and precision oncology. We also discuss current challenges, including data heterogeneity, model interpretability, reproducibility, and clinical translation, together with emerging strategies for integrating multimodal data such as radiomics and digital pathology. Finally, we introduce precision phenomics as a unifying framework that links molecular, spatial, functional, and clinical characteristics of tumors to support personalized cancer management. Collectively, AI-driven multiomics integration has the potential to transform lung cancer research and improve patient outcomes.