Genomics, as the earliest and most established omics technology, serves a foundational role in the omics field. Genetic susceptibility of diseases and large-scale data analysis facilitate disease detection. With continuous advancements in genomics technology and understanding of the pathogenesis of ALI/ARDS, numerous genes associated with risk and severity of ALI/ARDS have been identified (84), such as angiopoietin-2 (85) and pre-elafin (PI3) (86). These genes, which are associated with susceptibility to ARDS, provide understanding of its pathogenesis and contribute to the identification of new biomarkers and therapeutic targets. Gene-based therapy is becoming more prominent in clinical practice (87-89). In recent years, studies have implicated ferroptosis in the pathogenesis of ALI/ARDS by promoting the accumulation of reactive oxygen species (90,91). Studies (54,59) have reported genetic targets using both candidate gene and genome-wide association study (GWAS) approaches, including nicotinamide phosphoribosyl transferase, toll-like receptor 4 (TLR4), myosin light-chain kinase (MYLK) and macrophage migration inhibitory factor (59). Genomic methods such as DNA microarray and reverse transcription (RT)-PCR confirmed that the inhibitor of apoptosis-stimulating protein of p53 protects against ALI by inhibiting ferroptosis, thereby mitigating oxidative stress via the nuclear factor erythroid 2-related factor 2 (Nrf2)/hypoxia-inducible factor α/transferrin signaling pathway (54). These studies (67,92) offer insights and potential therapeutic avenues for clinical treatment (). Furthermore, through meta-analysis and GWAS of ARDS, Guo et al (67) identified a novel susceptibility locus with two annotated genes BLOC-1 Related Complex Subunit 5 and Dual specificity phosphatase 16. These genes are involved in immune-inflammatory processes and serve as functional genetic predictors of susceptibility to ARDS.

In addition, the mechanism by which microRNAs (miRNAs or miRs) contained in exosomes participate in intercellular communication and serve an immunomodulatory role in ALI/ARDS has investigated: Shen et al (93) found that miR-125b-5p is enriched in exosomes of adipose-derived stem cells (ADSCs) through miRNA microarrays and analysis of databases such as miRDB, miRtarbase, starBase and TargetScan (93). In Cecum Ligation And Puncture (CLP)-induced sepsis models (Six-to eight-week-old male healthy BALB/C mice), ADSC exosomes can alleviate lung tissue damage and reduce mortality rates by significantly increasing expression of Nrf2 and glutathione Peroxidase 4 (GPX4) (94,95). miR-125b-5p in ADSC exosomes can alleviate inflammation-induced pulmonary microvascular endothelial cell(PMVEC) ferroptosis by regulating the Keap1/Nrf2/GPX4 pathway, thereby improving ALI in sepsis (93). Consistently, Ma et al (96) synthesized a novel engineered exosome (N-exo) and found that N-exo containing miRNA 182-5p significantly improved ALI in vivo and in vitro by targeting the NADPH oxidase 4/Dynamin-related protein 1/NLRP3 signaling pathway (96). Additionally, Fan et al (97) used Illumina HiSeq 2500 to sequence benzo(a)pyrene (BaP)-associated miRNAs in plasma samples and screened differentially expressed miRNAs; hsa-miR-122-5p/Tumor protein 53 (TP53) axis regulates WNT5A through the non-canonical Wnt signaling pathway, thereby participating in BaP-induced lung epithelial injury (97). Wang et al identified and validated gene expression changes associated with pathogenesis of hemorrhagic shock-induced ALI/ARDS, including mRNA, miRNA, long non-coding (lnc)RNA and circular (circ)RNA, through high-throughput sequencing analysis and RT-quantitative (q)PCR; regulatory networks of lncRNA-miRNA-mRNA and circRNA-miRNA-mRNA may serve a key role in pathological processes, such as NOD-like receptor signaling pathway, JAK/STAT signaling pathway, cytokine/cytokine receptor interaction and so on (98). Whole exon sequencing refers to sequencing of all exons of a gene, although the exon region only accounts for ~1% of the whole genome and contains 85% of the disease-causing mutations. Therefore, whole exon sequencing has contributed toward understanding individual genomes to facilitate development of personalized treatment and prevention strategies. Compared with whole genome sequencing, its cost and technical requirements are lower (48). There have been recent reports on clinical cases using this technique to verify the metastatic origin of lymph node cancer cells in multiple types of primary lung cancer (99,100). In addition, Zhang et al (101) used whole exon sequencing to reveal diverse enriched pathways (such as p53/SLC7A11-cystine uptake-ferroptosis pathway and embryonic development pathway related to p53 and Mdm2 in normal lung tissue and pre-invasive and invasive adenocarcinoma (101). Whole exon sequencing is primarily employed in genetic diseases and rarely used in ALI/ARDS (50). However, with the maturity and popularity of this technology as well as combination of emerging technologies such as data mining, whole exon sequencing to provide personalized treatment for ALI/ARDS patients may be a possible application to facilitate future development of gene-targeted treatment.

Genomics facilitates understanding of the genetic heterogeneity and diversity of cells in patients with ALI/ARDS. It has been hypothesized that the ALI/ARDS susceptibility gene or promoter loci obtained by genomics can be inhibited or activated to prevent the occurrence and development of disease (51), suggesting a potential method to prevent or cure ALI/ARDS. However, the clinical and diagnostic applications of these identified genes (such as angiopoietin-2 gene (85) and PI3 (86)) a have insufficient specificities and sensitivities. Single nucleotide polymorphisms (SNPs) or single nucleotide structural variations have not been fully studied.

When compared with genomics, epigenomics provides a deeper understanding of the genetic mechanisms and processes of gene expression in ALI/ARDS. Epigenomics refers to investigation of gene activity regulation and expression changes that are not reliant on the DNA sequence (102). In ARDS, expression of cell-specific genes is distinguished based on their epigenomic features, such as DNA methylation, histone modification and local chromatin configuration (103). Epigenomics can reveal reversible and dynamic genetic changes and analyze epigenetic marks in different cell types to provide targets for the treatment of ALI/ARDS.

A growing number of studies are exploring epigenetic changes in ARDS, moving beyond traditional examination of the genome or transcriptome (69,104,105). DNA methylation, a key epigenetic modification, serves a pivotal role in ARDS pathogenesis by directly inhibiting transcription and expression by preventing specific transcription factors from binding their target sequences on candidate genes, thereby altering gene activity (106,107). DNA methylation variations may serve as specific diagnostic biomarkers for ARDS. Zhang et al (60) identified hypermethylation sites in peptidase inhibitor 3 (PI3) and microglial fractalkine receptor (CX3CR1) genes, which were associated with downregulated gene expression in ARDS, suggesting their involvement in pathophysiology. Additionally, a methylation profile study of patients with coronavirus disease 2019 (COVID)-19 and ARDS found hypomethylated probes linked to COVID-19 predictors, including genes participating in interferon regulation and viral response (70). This indicated increased gene expression during severe infection, highlighting potential biomarkers for ARDS in the context of COVID-19 (70). Further supporting the role of DNA methylation in ARDS, an integrative omics study identified the rs7967111 genetic variant at 12p13.2, which was significantly associated with increased risk of ARDS (38). Sensitivity analysis and pleiotropy evaluation suggest that this locus might act as a pathogenic ARDS mutation, interfering with the cascade of DNA methylation and histone modification (38). Moreover, in a lipopolysaccharide (LPS)-induced ALI mouse model, increased DNA methyltransferase levels were observed, along with increased 5-methylcytosine levels, indicating higher DNA methylation levels (17). Genome-wide epigenomic analysis of lung tissue has revealed that several inflammation-associated genes (such as IL-10, IL-1β and CXCR2/5/6) (108) are downregulated through methylation, further supporting the role of epigenetic alteration in the disease process.

DNA methylation and histone modifications shape gene expression patterns (109). Histone modifications influence DNA accessibility by altering local chromosomal structure or regulating the binding of effector proteins that control transcription (110). Wu et al (111) investigated the mechanism of lactate-regulated sepsis-induced ALI through omics technology. METTL3, a key modifying enzyme responsible for m6A, is upregulated in sepsis-associated multiple organ dysfunction (111). Lactate promotes METTL3 transcription by upregulating p303-mediated H3K18la in the METTL3 promoter region, which increases m6A modification to stabilize Acyl-CoA synthetase long-chain 4 (ACSL4) mRNA, inducing mitochondrial damage and subsequent induction of ferroptosis in alveolar epithelial cells (111). The regulatory pathway of histone modifications suggests that targeting modification sites may be an effective therapeutic approach to alleviate ALI/ARDS.

Given the relative ease of DNA isolation and stability of DNA methylation marks, epigenetic signatures hold promise as biomarkers for ARDS. As it has been confirmed that DNA methylation or histone modification at certain sites is related to the degree of ALI/ARDS, artificial DNA methylation regulation in the lung may have an inhibitory effect on progression of ALI/ARDS (104). Currently, epigenomics often only focuses on the effects of a single gene or pathway, while the disease mechanism of ALI/ARDS involves multi-factor interaction, which requires validation studies in this field (112-114). Occurrence and development of ALI/ARDS is a dynamic and rapid process. Although epigenetic modifications have different effects at different stages, dynamic and long-term follow-up studies lack explanation of the dynamic characteristics. In addition, the limited sample size may prevent reflection of the pathological process. Therefore, large-scale studies and long-term follow-up are required.

Transcriptomics aims to capture both coding and ncRNA and quantify gene expression heterogeneity among cells, lung tissue and organs. RNAs are messengers that function between DNA and proteins; RNAs in single cells convert genetic information stored in DNA into proteins (115). The transcriptome can serve as an indirect indicator of protein expression, suggesting genomic activity in real-time (116). Transcriptomics analysis involves comprehensive examination of modifications in gene expression, offering insight into the regulatory mechanisms governing key genes and pathways in disease progression.. Transcriptional profile of various types of cell (such as endothelial cells and epithelial cells) in ALI/ARDS can determine the prognosis and outcome of ALI/ARDS (117,118).

Through comparative gene expression analysis between patients with ALI/ARDS and healthy controls, several differentially expressed genes (DEGs) that are closely associated with the disease have been identified (76). Among these markers, certain genes and transcription factors (such as cytoskeleton associated protein 2 Gene (CKAP2), purinergic receptor P2Y, G-protein coupled, 14 (P2RY14), Retinol Binding Protein 2 (RBP2) and Thymidylate Synthetase (TYMS)) serve key roles in disease onset and progression; others serve as molecular markers associated with disease prognosis (76). These markers enhance understanding of ALI/ARDS pathogenesis and offer new avenues for early diagnosis, disease monitoring and the development of personalized treatments.

Transcriptomic technology has highlighted major alterations in immune dysfunction-associated gene expression in lung parenchyma. C-reactive protein and serum and amyloid A1 are among the most upregulated genes at both transcriptional and translational levels, making them potential predictive markers for diseases such as COVID-19 (66). Similarly, in patients with ALI/ARDS, the regulation of immune responses plays a key role in disease severity. Decreased number and proportion of regulatory T cells are associated with worsened disease outcome, while elevated levels of pro-inflammatory cytokines contribute to disease exacerbation. If the regulation process of transcription and expression of pro-inflammatory cytokines is inhibited, the severity of ALI/ARDS may be decreased, and imbalance of inflammatory response may be regulated to a balanced state so as to slow progress of ALI/ARDS (119,120). Furthermore, transcriptomics holds considerable promise for developing novel therapeutic strategies for ALI/ARDS. For example, Ma et al (90) used the Gene Expression Omnibus database to analyze ALI/ARDS models and identified immune-mediated ferroptosis genes involved in oxidative stress and metabolic processes, suggesting potential targets for immunotherapy aimed at modulating underlying pathophysiology. Transcriptome studies in lung tissue have also revealed that furosemide mitigates inflammatory responses and oxidative stress by modulating the thioredoxin-interacting protein/NOD-like receptor thermal protein domain-associated protein 3 (NLRP3)/gasdermin D pathway, influencing inflammasome activation and pyroptosis (121-123). Wang et al performed high-throughput analysis of hemorrhagic shock-induced ALI/ARDS, identifying a network of lncRNA-miRNA-mRNA (including 12 lncRNAs, 5 miRNAs and 15 mRNAs) and circRNA-miRNA-mRNA interactions (including 16 miRNAs and 39 mRNAs and 10 circRNAs) associated with ALI/ARDS (98). The discovery of these regulatory changes facilitates development of novel treatments targeting the metabolic processes in ALI/ARDS. miRNA has become one of the therapeutic strategies of choice. Downregulated the aforementioned miRNAs (such as MAP2K3, Copine 1 and Cortactin(Map2k3, Cpne1, and Cttn)) may serve as a treatment of ALI/ARDS.

The large amount of data involved in transcriptomic techniques presents difficulties in processing and analysis. The standardization of data processing and analysis is Standardization of data processing and analysis is challenging due to the need to apply different experimental methods. Therefore, development of efficient data processing tools and algorithms and the exploration of novel data integration and standardization methods are required. Advancing the integration of transcriptomic techniques and other omics techniques to understand the function and regulatory networks of biological systems can facilitate further research. Moreover, the clinical implementation of transcriptomic markers necessitates further validation and optimization to ensure accuracy and reliability.

Compared with traditional cell population analysis, transcriptomic data offer a novel perspective on understanding gene expression, revealing regulatory mechanisms of key genes and pathways involved in disease processes (124). Advances in cell capture, phenotyping, molecular biology and bioinformatics continue to expand the potential for transcriptomics in biological and medical applications (125). Further technological progress and research may identify clinically relevant transcriptomic markers for ALI/ARDS to redefine treatment approaches, improving patient outcomes and quality of life.

Proteins are the primary building blocks of life and maintain most cellular functions, including replicating DNA, facilitating transduction of genetic information, catalyzing metabolic reactions and driving cellular motility (126). Because protein profiles reflect organ-specific information more accurately than DNA or RNA alone, proteomics may reveal mechanisms that cannot be identified at the genomic level and are directly associated with clinical phenomena. Deep proteome profiling has identified biomarkers linked to ALI/ARDS progression, such as caveolin-1 (127), MMPs (128), vascular endothelial growth factor (VEGF) (129), receptor of advanced glycation end products (130), tumor necrosis factor (TNF-α) and IL-8 (131), providing insights into diagnosis, prognosis and mechanistic understanding of the disease at all stages (132). In 2003, proteomics was proposed to analyze altered protein expression comprehensively, allowing early biomarker discovery, refined therapeutic strategies and accelerated drug development (133). Increasingly, proteomics studies use diverse sample types, such as BALF (134), lung tissue (135), blood and exhaled breath condensate (136), offering flexibility in identification of key proteins and pathways associated with ALI/ARDS. This approach enhances understanding of inflammatory and repair signaling pathways and specific proteins relevant to ALI/ARDS pathology with potential clinical applications as biomarkers.

ALI/ARDS triggers complex changes in lung protein expression; >100 protein biomarkers have emerged as potential therapeutic targets (137). In 2004, Bowler et al (138) profiled proteins in plasma and pulmonary edema fluid of patients with ALI, identifying >300 distinct proteins (138). These biomarkers, analyzed through methods such as protein-protein interaction networks, data-independent acquisition proteomics and parallel-reaction monitoring, demonstrate promising diagnostic and prognostic potential (19). Studies have highlighted the involvement of proteins such as caveolin-1 (127), MMPs (128), vascular endothelial growth factor (VEGF) (129), receptor of advanced glycation end products (130), tumor necrosis factor (TNF-α) and IL-8 (131). These proteins are primarily expressed by alveolar epithelial or endothelial cells. Therefore, regulation of protein expression in epithelial or endothelial cells during ALI/ARDS may be a strategy for future treatment. However, a holistic approach is needed to capture the dynamic, multifaceted changes that characterize ALI.

Recent advances in mass spectrometry (MS) technology have enabled high-throughput protein sequencing, leading to broader application of proteomics in ALI/ARDS research (139,140). MS-based proteomics techniques include isotope-encoded affinity tagging, stable isotope labeling of amino acids in cell culture, isobaric tagging for relative and absolute quantification (iTRAQ), gas chromatography and liquid chromatography (LC). Xu et al (56) used iTRAQ to identify proteins differentially expressed in endotoxin-induced porcine ARDS models, noting that IL-1 receptor antagonist protein, α-trypsin-interacting inhibitor heavy chain H4, MMP-1 and MMP-10 are highly expressed after LPS-induced ARDS and decreased following hemadsorption (HA) treatment. This indicates HA treatment could mitigate ARDS by curbing cytokine storms, improving alveolar barrier integrity and restoring proteomic balance in the exudative phase. Clinical verification is required to determine whether HA can improve disease progression of patients with ALI/ARDS (56). Adrover et al (61) used LC-MS to reveal that a CXCR2- and circadian-regulated program in neutrophils decreases inflammation by modulating proteome profiles. This depletes granule content and extracellular trap formation, leading to a reduction in inflammatory proteins (61). Dong et al (68) conducted proteomic assay and Mendelian randomization to identify plasma insulin-like growth factor binding protein 7 as a novel biomarker involved in platelet function in ARDS pathology, providing a promising target for further experimental and clinical research.

Proteomics methods are relatively complex and certain proteins may exhibit instability, which limits their application. Thus, optimizing proteomic analysis methods may be a future research focus. Additionally, proteomics can be integrated with other omics techniques to mitigate its limitations, thereby advancing understanding of complex biological systems such as ALI/ARDS.

The application of proteomics has identified protein biomarkers for ALI/ARDS diagnosis and prognosis (such as fatty acid-binding protein 5 (67), Chromosome Segregation 1 Like) (141). However, the human proteome is dynamic and certain protein biomarkers exhibit structural diversity after modification. PTMs are chemical modifications that are primarily catalyzed by enzymes that play key roles in the functional proteome and are the key mechanism underlying increased proteomic diversity (142). PTMs are commonly involved in complex and dynamic cellular processes by regulating cellular activity, localization and interactions with other cell molecules such as proteins, nucleic acids, lipids and cofactors (143). PTMs include phosphorylation, glycosylation, ubiquitination, methylation, acetylation, lipidation and protein hydrolysis, and are involved in almost all aspects of normal cell biology and pathogenesis (144). Moreover, there is a growing body of evidence implicating PTMs in a number of human disease mechanisms, which makes the utilization and understanding of PTMs key in the discovery of molecular mechanisms and therapeutic targets for ALI/ARDS (144,145).

Protein phosphorylation is one of the most studied and common PTMs, occurring primarily on threonine, serine, and tyrosine residues (146-148). Protein phosphorylation serves a major role in regulating the cell cycle, division, apoptosis and signal transduction as a key mechanism of cellular signal transduction (149). Zhao et al (150) demonstrated the importance of MAPK and the JAK/STAT pathway by enriching the transcriptome of lung tissue, followed by protein immunoblotting to demonstrate that Xie-Bai-San (XBS) inhibits phosphorylation of extracellular signal-regulated kinases (ERKs) and STAT3 to combat ALI (150). PTMs occur on different amino acid side chains or peptide bonds and are usually mediated by enzymatic activity (146). Li et al (151) found that histone deacetylase, an important epigenetic modifying enzyme, can promote ALI by upregulating the phosphorylation of Rho-associated protein kinase 1 (151).

In conclusion, protein phosphorylation and dephosphorylation are the most prevalent key regulatory mechanisms that regulate and control protein viability and function. Protein phosphorylation plays an important role in the pathological mechanisms of ALI/ARDS and may serve as a biomarker and potential therapeutic target for ALI/ARDS. However, most of the biomarkers in phosphoproteomics are still in the preclinical stage and there are only few methods available to detect protein phosphorylation with low specificity, such as mass spectrometry and quantitative phosphoproteomics (152). Therefore, there is need for larger prospective cohort studies on phosphoproteomics, as well as development of assays and reduction of assay cost to facilitate their application in clinical practices.

With the development of PTMs in recent years, lactate proteolysis has become a focus of research (146). Lactate is a key metabolite of glycolysis, and lactate proteolysis is crucial for lactate functioning, which is involved in glycolysis-related cellular function and macrophage polarization (153). Lactate proteolysis is a protein modification induced by lactate accumulation, which can affect gene transcription by altering spatial conformation of histones and thereby regulating the expression of associated genes (154). Increased production of lactate under hypoxia and mitochondrial dysfunction and decreased clearance due to renal and hepatic injury facilitate the accumulation of lactate in septic patients; therefore, high levels of serum lactate are a key biomarker of sepsis prognosis (155).

Lactation is involved in inducing hyperpermeability of endothelial cells. Fan et al (156) found that lactate causes vascular permeability and exacerbates organ dysfunction in CLP-induced sepsis. Lactic acid-induced ERK-dependent activation of calpain 1/2 for hydrolytic cleavage of vascular endothelial (VE)-calmodulin leads to enhanced endocytosis of VE-calmodulin in endothelial cells. Lactate-induced phosphorylation of ERK2 promotes the dissociation of ERK2 from VE-calmodulin. In vivo inhibition of lactate production or gene depletion of the lactate receptor G protein-coupled receptor 81 (GPR81) attenuates vascular permeability and multiorgan injury and improves survival outcomes in polymicrobial sepsis. Overexpression of heat shock protein A12B (HSPA12B) prevents lactate-induced disassembly of VE-calmodulin (156). Lactate promotes vascular permeability by decreasing expression of VE-calmodulin and tight junction proteins, and the deleterious effects of lactate on vascular hyperpermeability are mediated by HSPA12B- and GPR81 (a specific receptor for lactate)-dependent signaling (156). Lactation is involved in the regulation of macrophage polarization: Yao and Yang (157) revealed the involvement of lactation in the regulation of gene expression during M1 macrophage polarization (157). Wang et al (158) revealed that lactic acid enhances the lactation of pyruvate kinase isozyme typeM2 and promotes the transition of pro-inflammatory macrophages to a reparative phenotype. Further, Dichtl et al (159) suggested that lactylation could be a consequence of macrophage activation rather than its cause. However, Wang et al (160) revealed that proline lactylation promoted M1 macrophage polarization which in turn reduces lactic acid.. Thus, macrophage polarization is important in numerous diseases, and the association between lactation and macrophage activation warrants further in-depth study (154).

Ubiquitination is a key PTM mediated by three enzymes (E1, E2 and E3) that are involved in the degradation of proteins as well as in the regulation of cellular functions, such as cell cycle, proliferation, apoptosis, differentiation, metastasis, gene expression, transcriptional regulation, signaling, repair of damage and inflammatory immunity (161). Ubiquitination serves an important role in the pathogenesis of ALI and other types of lung diseases, and the expressions of various inflammatory and anti-inflammatory factors are regulated by ubiquitination. For example, the deubiquitinating enzyme Ubiquitin Specific Peptidase 13 (USP13) stabilizes the anti-inflammatory receptor immunoglobin interleukin-1-related receptor (IL-1R8/Sigirr) to inhibit lung inflammation (162). Qian et al (163) found E3 ubiquitin ligase, tripartite motif-containing 47 (TRIM47), is highly expressed in vascular endothelial cells and the knockdown of TRIM47 inhibits the transcription of pro-inflammatory cytokines, thereby suppressing LPS-induced lung inflammation and ALI (163). Li et al (164) found that inflammatory cytokine secretion is decreased by inhibiting Nrf2 ubiquitinated proteasomal degradation, which attenuates LPS-induced ALI (164). There is growing evidence of the involvement of ubiquitination in development of ALI/ARDS, but studies of ALI/ARDS based on ubiquitination proteomics are not comprehensive (165,166). Optimization of ubiquitination proteomics methodology and may identify ubiquitination biomarkers.

Protein phosphorylation, endocytosis and ubiquitination are the most prevalent key mechanisms that regulate and control protein viability and functions. Protein phosphorylation, endocytosis and ubiquitination serve key roles in the pathological mechanisms of ALI/ARDS and may serve as biomarkers and potential therapeutic targets. However, most biomarkers for modified proteomics are still in the preclinical stage, with few assays and low specificity. Therefore, there is need for larger prospective cohort studies of modified proteomics, as well as development of additional assays and decreased costs to facilitate application in clinical practice.

Metabolomics involves qualitative and quantitative analyses of low-molecular weight metabolites or small molecule chemicals involved in metabolism and present in an organism or cell to reveal the association between dynamic metabolic changes and pathophysiological processes influenced by internal and external factors (167). Metabolomics uses information modeling and system integration to evaluate indicators, thereby offering insights into dynamic metabolic changes in organisms. Metabolomics applications involve techniques to elucidate large differences in polarity, charge and size of multiple metabolites, and the most common techniques are based on nuclear magnetic resonance spectroscopy and MS (168). Metabolomics can analyze changes in cell metabolic state in real-time and reveal the metabolic pathways in development of ALI/ARDS to facilitate the discovery of novel biomarkers for early diagnosis and prognosis evaluation.

Moreover, advancements in MS technology have enhanced the speed of small molecule analysis in cell metabolomics, enabling high-throughput metabolomics and novel routes for systems biology, functional genomics, drug discovery and personalized medicine (169,170). For example, using an LC-MS platform, Evans et al (171) analyzed BALF samples from patients with ARDS and healthy controls, identifying 37 biomarkers primarily affecting amino acid metabolism, glycolysis, gluconeogenesis, fatty acid biosynthesis and phospholipid and purine metabolism. Similarly, Nan et al (71) used ultra-high-performance LC-MS/MS to examine the metabolomic profiles of patients with community-acquired pneumonia (CAP). The aforementioned study distinguished acute and remission phases of CAP and identified myristoyl lysophosphatidylcholine (LPC) in patient plasma, a metabolite inversely associated with disease severity, suggesting its potential as a biomarker and therapeutic target for pneumonia. Additionally, Fan et al (172) used metabolomics to differentiate and diagnose ALI in patients with acute aortic dissection, identifying β-hydroxybutyrate and TNF-α as independent risk factors for ALI. The aforementioned study also highlighted elevated levels of pyruvate, alanine, malondialdehyde and lactic acid, along with changes in amino acid profiles, providing valuable insight into ALI mechanisms and potential biomarkers for diagnosis and treatment (172).

Large-scale metabolomics research is increasingly used to explore individual responses to environmental stimuli based on genetic background and time-dependent variations (173,174). Beyond observing metabolic shifts following drug administration, metabolomics offers a detailed and accurate view of drug mechanisms and therapeutic efficacy. Li et al (62) employed Ultra Performance Liquid Chromatography (UPLC-triple-time of flight (TOF)/MS to analyze the metabolic effects of Ganoderma atrum polysaccharide (PSG) on rats; significant alterations in histidine, nitrogen, tryptophan and glycerol phospholipid metabolism were observed, which supported the protective effect of PSG against ALI, preserving lung histology and decreasing pro-inflammatory cytokines and NO in serum and lung tissue. Similarly, using Ultra Performance Liquid Chromatography Quadrupole TOF-MS (UPLC-QTOF-MS), Hu et al (63) examined plasma samples from LPS-induced ALI rats treated with crude Scutellariae radix (CSR) and wine-processed S. radix (WSR) and found 16 biomarkers in LPS-induced rat plasma; CSR influenced ALI by regulating abnormal sphingolipid metabolic pathways, whereas WSR-treated samples primarily showed adjustments in retinol and tryptophan metabolism, indicating mechanisms of these treatments.

However, metabolomics research has problems. For example, small sample sizes may prevent detection of significant differences. Differences in the metabolome of patients with ALI/ARDS caused by different diseases may affect the further report of metabolomics. For metabolomics to become a reliable tool in intensive care, issues associated with sample collection and processing, multivariate data analysis and patient selection must be addressed (175). For example, patients tend to choose traditional clinical application techniques and there is instability of metabolic markers (176). Future research should improve speed and precision of metabolomics in detecting metabolites across large samples. Developing methods for accurately analyzing small samples is key for expanding metabolomic clinical utility.

Gene expression profiling through genomic or transcriptomic sequencing serves a central role in understanding disease genetics (178,179). However, relying solely on genomics or transcriptomics limits insight due to individual variations in DEGs across populations and lack of clear gene-disease associations. Integrating genome and transcriptome data provides a more comprehensive approach to investigating complex diseases, bridging genetic variations and disease phenotypes.

Combining genomics and transcriptomics has advanced understanding of the molecular mechanisms driving ALI/ARDS, especially pathways involved in inflammation (180). Furthermore, by examining gene expression dynamics, integrated genomics and transcriptomics offer valuable insight into potential targets for disease treatment. Compared to using one of these technologies alone, transcriptomics and genomics are more likely to reveal responses and expression pathways of specific cell populations at the cell level. The increased understanding of inflammatory pathways at the cellular level is more conducive to implementation of therapeutic strategies for specific cell types, such as lymphocytes, monocytes and endothelial cells (181). Jiang et al (65) used single-cell RNA sequencing (scRNA-seq) and differential gene analysis to compare peripheral blood mononuclear cells from patients with pneumonia and sepsis with early ARDS with those without ARDS; downregulated cytokine signaling 3 (SOCS3) expression in monocytes from patients with ARDS was observed, alongside increased expression of IFN-induced pro-inflammatory genes (such as Ras-related protein in brain 11A (RAB11A)), which may inhibit exocytosis of neutrophils, particularly in CD16⁺ cells, suggesting that monocytes in patients with ARDS exhibit dysregulated gene expression patterns beyond inflammation-associated genes. This finding may suggest that improving the monocyte intrinsic regulatory system can alleviate ALI/ARDS (65). Other studies have demonstrated the use of integrated genomics and transcriptomics in understanding cell-specific responses (57,75,80,89,182). Lai et al (75) performed scRNA-Seq to study expression of the protein arginine methyltransferase 4 (PRMT4) gene, known to mediate lymphocyte apoptosis, in activated T lymphocytes, finding its upregulation induces caspase 3-mediated cell death signaling. Short hairpin (sh)RNA-mediated PRMT4 knockdown inhibits LPS-induced caspase 3 activation, suggesting it as a promising therapeutic target. Huang et al (57), through multi-network analysis, explored how Kruppel-like factor 2 (KLF2) regulates endothelial integrity in ALI. Overexpressed KLF1, a regulator of ARDS-associated genes associated with pulmonary microvascular endothelial cytokine storm, oxidation and coagulation, was found to improve LPS-induced ALI. Conversely, expression of pulmonary KLF2, a key ARDS-associated gene regulator, decreased. KLF2 improves endothelial barrier function and activates the Ras-related C3 botulinum toxin substrate 1 (Rac1) pathway in human microvascular endothelial cells via the Rap guanine nucleotide exchange factor 3/exchange protein directly activated by cAMP 1 pathway. Furthermore, DEG analysis and scRNA-Seq in a mouse malaria-associated ARDS model revealed an increase in endothelial cell number during disease regression. Therefore, promoting the proliferation of endothelial cells may be an interesting novel treatment for malaria-associated ARDS in combination with antimalarial drugs (80). MultiNicheNet interaction analysis has unveiled substantial modifications in critical ligand-receptor interactions during ARDS remission, offering potential targets for novel therapeutic strategies (80). Clinical studies using mRNA and miRNA genomics and transcriptomics show T cell dysfunction in patients with ARDS, highlighting the therapeutic importance of regulating immune pathways (81,182).

Combined analysis of transcriptomics and genomics enhances ability to diagnose and treat ALI/ARDS. Nonetheless, further exploration of specific molecular mechanisms is needed to realize this potential.

In the complex pathology of ALI/ARDS, finding novel diagnostic markers is key (183). Compared to using one of these technologies alone, integrating transcriptomic and epigenomic analyses provides deeper insight into disease pathogenesis and potential therapeutic targets (184,185). In ALI/ARDS, transcriptomics helps identify gene expression changes and functional pathways, define gene function and map regulatory networks (186). However, transcriptome data alone are not sufficient to reveal the disease complexity. Unlike the genome, which solely contains genetic information, the epigenome expresses genetic information through modifications. Because cis-acting elements cannot encode proteins, trans-acting elements such as external transcription factors (TFs) and epigenetic modifications serve key regulatory roles in gene expression. Epigenomics provides additional context by capturing gene regulation via the aforementioned nc elements. Thus, compared to using one of these technologies alone, combining these data allows a more comprehensive view of transcriptional regulation, including pathway activity and transcription factor interactions (187).

Integrating transcriptome and epigenome data is valuable for understanding the association between ALI/ARDS and genetic polymorphisms. The identification of genetic polymorphisms can identify increased risk of disease within a certain group. Tejera et al (188) investigated the association between genetic variants and ARDS susceptibility in a cohort of 449 patients with ARDS and 1,031 at-risk patients who did not meet the criteria for ARDS during hospitalization and genotyped NPs, particularly the variant rs2664581; ELISA of plasma PI3 levels revealed that rs2664581 (or associated SNPs in linkage disequilibrium) may impact PI3 capacity to inhibit heightened human neutrophil Elastase activity, a genetic variant linked to ARDS risk. Notably, different rs2664581 genotypes resulted in varying plasma PI3 levels, with the Hap2 (TTC) haplotype, containing the variant allele rs2664581, identified as a risk factor for ARDS. Moreover, plasma PI3 level was significantly lower in patients with ARDS with a allele homozygosity than in those with the C allele variant. These findings underscore the association between PI3 gene polymorphisms, ARDS susceptibility and circulating plasma PI3 levels, thereby emphasizing the role of genetic and epigenetic factors in disease risk and highlighting PI3 as a potential biomarker for targeted therapeutic intervention.

Integration of transcriptome and epigenome analysis offers deeper understanding of ARDS pathogenesis to unveil new potential inflammatory targets for its prevention and treatment. Sun et al (189) expanded on previous findings (190) linking variants of MYLK gene to severe asthma susceptibility by investigating the role of the non-muscle (nm) MYLK promoter region in ARDS. The 2512-bp DNA of this region was synthesized based on the NM_053025 sequence. Site-directed mutagenesis was used to modify the nmMYLK DNA sequence, which was sequenced to examine its function in ARDS. The aforementioned study also conducted electrophoretic mobility shift assays to detect TF binding to nmMYLK and tested promoter demethylation using 5-aza-2′-deoxycytidine in an LPS-induced ALI mouse model to verify the role nmMYLK. The aforementioned study revealed that exogenous and endogenous inflammatory factors, such as cytokines, mechanical stress, hypoxia and DNA demethylation, can significantly increase nmMYLK promoter activity, influencing ARDS-associated SNPs by altering TF binding and increasing nmMLYK expression, potentially affecting inflammatory severity in ARDS (189). Increasing activity of this site may serve a crucial role in suppression of inflammation in clinical ALI/ARDS cases.

Integrating transcriptome and epigenome data helps clarify the role of plasma cytokine levels. Kovacs-Kasa et al (191) observed elevated plasma cytokine levels in severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2)-infected patients but these cytokines alone did not increase pulmonary microvascular endothelial permeability directly. The plasma factors causing pulmonary microvascular injury and increased endothelial permeability remain unknown. Electric cell-substrate impedance sensing assessment of plasma factors demonstrates that thrombin, angiopoietin 2, VEGF, complement factors C3a and C5a and spike protein contributed to increased endothelial permeability, although SARS-CoV-2 plasma induces milder, shorter-lived endothelial injury compared with plasma from patients with unconfirmed SARS-CoV2 infection. Among 15 cytokines analyzed, plasma levels of the pro-inflammatory cytokines IL-1b, IL-2, IL-6, IL-8 and IL-17 are elevated in patients with novel CoV, with IL-10 exhibiting the greatest increase compared with the normal control group (191). Pretreatment with thrombin inhibitors, neutralizing antibodies against cytokines, C3a and C5a receptor antagonists or angiotensin-converting enzyme 2 (ACE2) antibodies alone or in combination does not significantly decrease SARS-CoV-2-induced endothelial permeability, underscoring the complexity of SARS-CoV-2-associated endothelial injury (192).

Integrating transcriptome and epigenome analyses is instrumental in identifying drug targets and underlying mechanisms. Xian et al (193) combined shRNA lentivirus silencing, RNA isolation, qPCR, immunofluorescence, confocal microscopy and chromatin immunoprecipitation to investigate the effect of metformin on ARDS; the primary molecular target, Electron Transport Chain Complex I (ETCCI), exerts an inhibitory effect by decreasing ATP production. Metformin prevents NLRP3 inflammasome activation, macrophage infiltration and collagen deposition by inhibiting TLR-induced mitochondrial DNA (mtDNA) synthesis (193). These results indicate that metformin, an affordable and well-tolerated drug, may be repurposed to prevent and alleviate LPS-induced ARDS, highlighting it as a potential therapeutic agent.

In summary, integrating transcriptomics and epigenomics facilitates understanding of the association between gene polymorphisms, changes in plasma cytokine levels and gene modifications in ARDS. These insights are key for the future of disease diagnosis, treatment and prognosis assessment. However, challenges remain, including the need for advanced technology and equipment for data acquisition and processing, ongoing optimization and updating of bioinformatics methods and the translation of research findings into applications for clinical practice.

Transcriptomics involves comprehensive analysis of the entire transcriptional activity within cells at specific developmental or physiological stages, allowing for the detection and quantification of specific nucleic acid sequences (194). By contrast, proteomics is the systematic study of all proteins expressed within a genome to unravel the molecular networks governing complex biological processes at the protein level (139). Integrating transcriptomic and proteomic data offers holistic understanding of biological samples, providing critical insight into physiological processes and disease mechanisms. In ALI/ARDS research, this combined approach has aided in understanding disease pathogenesis and identifying potential therapeutic targets.

As a key reaction in ALI/ARDS, inflammatory response plays an important role in progression and prognosis: Activation of inflammatory responses aggravates ALI/ARDS severity (9,77,195). Understanding the molecular mechanisms underlying the inflammatory response holds promise for improving diagnosis and treatment of ALI/ARDS. Considering the heterogeneity and limited research findings (196), DNA microarray analysis and genotyping have been integrated with proteomics to study ALI pathogenesis. Tang et al (197) recently analyzed the inflammatory response characteristics of ALI/ARDS in obesity by targeting the transcriptional profile of lung tissue in high-fat-induced obese ALI mice. The biological functions of DEGs and expression characteristics of protein were explored through transcription profiling, Gene Ontology and Kyoto Encyclopedia of Genes and Genomes analysis. The comprehensive transcriptome and proteomics verified that lncFirre is an effective therapeutic target for obesity-associated ALI and revealed its inflammatory signaling mechanism in ALI. Moreover, inflammation-associated lncRNAs played a crucial role in the inflammatory pathway of ALI, which provides precise targets for drug development (198). Furthermore, by merging proteomic with transcriptomic data, studies have identified key tissue and circulating biomarkers that predict disease progression or deterioration in patients with idiopathic pulmonary fibrosis and ALI (199-201). Using a combination of RNA-seq and SomaLogic proteome assay, significant increases in pathways such as mast cell chemokines, T helper-2 (Th-2) axis and Wnt signaling were identified, providing valuable diagnostic insights for ALI/ARDS (199). Sinha et al (200) employed scRNA-Seq and plasma proteomics to investigate pathological mechanisms underlying ARDS. The aforementioned study characterized a COVID-19-enriched neutrophil state, its mechanism of action and dexamethasone-mediated modulation of neutrophil function to support the development of targeted immunotherapies for severe COVID-19. It remains unclear whether dexamethasone improves ARDS caused by COVID-19 in a manner similar to that in patients with ARDS from other causes. Moreover, by applying scRNA-Seq and unbiased serum proteomics, Ramaswamy et al (201) investigated multisystem inflammatory syndrome in children and COVID-19 in adults and healthy individuals, revealing specific alterations, such as heightened cytotoxic gene expression in natural killer and CD8⁺ T cells, which could aid in early disease diagnosis and prognosis assessment. Thus, transcriptomics and proteomics are pivotal in examining immune response-associated characteristics and mechanisms in ARDS progression.

Oxidative stress, a key factor in ALI/ARDS pathogenesis, has been studied using transcriptomic and proteomic tools. Application of quantitative redox proteomics has revealed that S-glutathione promotes interaction between fatty acid binding protein 5 and peroxisome proliferator-activated receptor (PPAR)β/δ, activating PPARβ/δ target genes and suppressing LPS-induced macrophage inflammation (67). Additionally, a comparison of ALI mouse models underscores the key role of S-glutathione in cellular antioxidant defense during ALI (67). MMPs serve as key mediators and effectors of alveolar-capillary membrane damage and repair due to oxidative stress in ALI/ARDS. They contribute to degradation of non-structural extracellular matrix (ECM) components in experimental lung injury, including syndecan-1/keratinocyte-derived chemokine and macrophage inflammatory protein-1α (128). Further studies have highlighted Cav-1 as a regulatory factor in apoptotic pathways in different stages of lung injury (202,203). Cav-1 interacts with or is upregulated by death receptor agonists such as Fas ligands, TNF and TNF-related Apoptosis Inducing Ligand. Under normal conditions, it interacts with the key autophagy marker protein light chain 3B, a relationship that is disrupted by reactive oxygen species (ROS). This suggests Cav-1 is involved in regulating autophagy-mediated cell death during ALI/ARDS (204-206). In summary, a combination of transcriptomics and proteomics can assess the extent of oxidative stress. Targets of oxidative stress are expected to provide a new way for the treatment of ALI/ARDS.

The potential of single-cell multi-omics technology in diagnosing, assessing treatment strategies, predicting prognosis and outcomes of ALI/ARDS continues to expand (207). Integrating metabolomics with other omics data has become essential in understanding diseases and developing effective treatments (172,175,208). Analyzing the association between DEGs obtained from transcriptome data with differential metabolites can examine organismal variations at causal and effectual levels. This combined approach allows identification of pivotal genes, metabolites, metabolic pathways and regulatory networks that underlie complex disease mechanisms. Additionally, genomics aids in identifying ALI/ARDS susceptibility-associated genetic targets (209,210). Single-cell multi-omics techniques may overcome the inherent limitations of individual omics approaches, providing more precise, reliable data for comprehensive understanding of disease mechanisms.

Single-cell multi-omics technology has recently made notable contributions to the study of ALI/ARDS and other diseases, especially in elucidating cellular and organelle-level molecular mechanisms in ALI/ARDS (207,211). Zhang et al (212) used such as glucose uptake and mitochondrial calcium measurement, fluorescence staining, protein blot analysis, apoptosis assay, ROS flow cytometry analysis, small interfering RNA duplex transfection, total RNA isolation and real-time RT-PCR to explore the mitochondrial pathway involved in hyperoxia-induced lung injury, which revealed a novel TLR4/stanniocalcin 1 (STC1)-mediated mitochondrial pathway that serves both homeostatic and oxidant-induced cytoprotective roles in lung endothelial cells. Similarly, Wang et al (213) explored the potential molecular mechanism of AU-rich binding factor 1 (AUF1, in regulating iron homeostasis in sepsis-induced ALI: Co- and RNA immunoprecipitation and measurements of cell viability, lipid peroxidation, iron accumulation and total glutathione levels demonstrated that activating the AUF1 pathway may alleviate sepsis-induced ALI. Single-cell multi-omics has demonstrated elevated iron and Malondialdehyde levels in lung tissue, along with marked NRF2 downregulation and activating transcription factor 3 (ATF3) upregulation, further validating AUF1 therapeutic potential in alleviating CLP-induced ALI.

Exosomes have promise in mitigating ALI/ARDS (78,79,214). Derived from human mesenchymal stem cells and other sources, exosomes have been studied for therapeutic effects on macrophage function and inflammatory response in LPS-induced ALI (78,79,214). RNA extraction, RT-qPCR, western blotting and ELISA have demonstrated that human mesenchymal stem cell-) derived exosomes (hMSC-exos) derived exosomes effectively attenuate glycolysis and sepsis-triggered inflammatory responses in macrophages, thereby decreasing lung damage (215). Adipose-derived exosomes, a promising therapy for ALI, are potent protectors against LPS-induced ALI in mice. Alongside exosomes, inflammasomes are key in ALI/ARDS progression (216). A 4-benene-indol derivative significantly inhibits NLRP3 inflammasome activation in LPS-induced ALI, suggesting a therapeutic target for sepsis-related ALI (217). By performing RNA isolation, real-time qPCR, flow cytometry, immunoprecipitation and protein blot analysis, Xie et al (218) identified High mobility group box 1 (HMGB1) as a factor that impedes macrophage-mediated exocytosis and prolongs inflammation by inhibiting member RAS oncogene family (Rab43)-regulated anterograde transport of CD91, indicating that restoring Rab43 may be beneficial in reducing inflammation and mitigating ALI/ARDS in humans (218). In a study involving patients with severe COVID-19, CD163-expressing monocyte-derived macrophages acquired a pro-fibrotic transcriptional phenotype during COVID-19 ARDS (69). Gene set enrichment and computational data integration revealed significant similarities between COVID-19-associated macrophages and profibrotic macrophage populations in idiopathic pulmonary fibrosis. This provides a deeper understanding of the ALI/ARDS mechanism and potential therapies (69). The aforementioned comprehensive investigations used quantitative proteomics, immunohistochemistry, scRNA-seq, clinical CT imaging and clinical evaluation to analyze transcriptional profiles of lung tissue cells, highlighting macrophage accumulation, monocyte-derived macrophage fibrosis and robust intercellular signaling in the ALI/ARDS pathophysiology. In addition, analysis of transcriptional and proteomic profiles in COVID-19-associated macrophages revealed that lung macrophages in COVID-19 express a range of scavenger receptors and proteins involved in exocytosis, including macrophage mannose receptor (MRC1), CD163, among others, as well as high levels of transforming growth factor-β 1 (TGFB1) and Transforming Growth Factor Beta Induced (TGFBI). These genes may directly or indirectly contribute to the profibrotic function of macrophages. Moreover, integrative study not only detailed cell-specific responses but also suggested that the presence of CD163/(Legumain) LGMN-Mϕ during convalescence may be the key to unraveling the mechanisms of fibrosis regression and reversing lung injury (69). Further exploration is warranted to elucidate the molecular mechanisms underlying virus- and macrophage-induced fibrosis in vivo.

Despite their advantages, omics techniques currently require cell isolation from tissue, which can disrupt intercellular interactions. However, advancements in spatial multi-omics hold promise for overcoming these limitations by allowing spatially resolved understanding of molecular interactions within tissues (207).

With continuous advancements in biotechnology, multi-omics technology may provide more precise characterization of the heterogeneity and prognosis of disease. Multi-omics technology and AI algorithms promote the development of precision medicine (116). Large-scale data from multi-omics technology is difficult to integrate (219). Poirion et al (220) proposed a unique computational modeling framework, DeepProg, which uses patient survival as the target model to predict novel patient survival risk. DeepProg is run by processing multiple types of omics dataset through a combination of deep learning and machine learning algorithms. This approach has advantages of high sensitivity and high specificity in cancer classification (221), survival characteristics (222) and treatment planning (223). The high predictability based on deep learning is primarily attributed to its ability to automatically capture non-linearities and perform dimensionality reduction and hierarchical representation. In addition, Cui et al (224) constructed a basic model, scGPT, that can reveal complex biological interactions in the single-cell domain by harnessing the power of pretrained transformers on a large amount of single-cell data. The data entered in the model in advance by the developer can then be transferred to downstream tasks through fine-tuning, such as cell type annotation, perturbation prediction and multi-batch and multi-group integration. The pretrained model demonstrates strong extrapolation capability to present meaningful clustering patterns on unseen datasets (138).

In summary, AI-based data integration studies of multi-omics data are primarily classified into the following four categories: i) Feature selection and dimensionality reduction, ii) prediction of clinical outcomes; iii) clustering for subtype discovery (225). High-dimensional multi-omics techniques analyze data integration through algorithms to observe molecular-molecular interactions and biological phenomena in living organisms. To the best of our knowledge, however, there are relatively few studies on AI algorithms in ALI/ARDS disease prediction and cell subtype classification (204,205). The use of high-accuracy prediction and drug utilization models may improve disease prognosis and drug selection for ALI/ARDS as well as the development of precision medicine.

There is extensive research on the application of multi-omics technology in ALI/ARDS (37,226,227). Application of multi-omics techniques facilitate diagnosis of diseases, such as through the upregulation or downregulation of serum markers mentioned above. In addition, multi-omics technology can clarify the disease mechanism of ALI/ARDS, such as the upstream and downstream communication between inflammatory response and reaction processes of oxidative stress (228). To translate current research into the clinical setting, biomarkers must be validated. The analysis of potential biomarkers by high-throughput omics technology requires collection of biological samples from patients at different stages, followed by verification of feasibility and effectiveness in clinical trials. The genomic data of patients should be integrated and analyzed to construct a patient information database platform for ALI/ARDS, identify multi-omics data associated with therapeutic drugs, and adjust personalized treatment plans for patients at different stages. A good feedback mechanism should be established. When feasible treatment schemes or diagnostic methods in research are applied to clinical practice, timely feedback to researchers and continuous exploration and development of basic experimental mechanisms is key for development of long-term and effective clinical solutions. Unbiased analysis of preclinical and clinical samples should be performed. Unbiased analysis should involve careful selection of data, ensuring that the sample is representative of the population, and that the analysis does not overemphasize certain subsets of data that may lead to skewed or inaccurate results. While researchers conduct animal experiments to complete preclinical research, the use of clinical ALI/ARDS samples to correct the problems which found in animal experiments can promote clinical research progress. Combination of AI and multi-omics can determine new solutions to clinical problems through secondary analysis of datasets. Classification and regression tree and artificial neural networks have been applied in individualized treatment of ARDS and the integration of algorithms can be more helpful than existing single algorithmic approach in clinical treatment (229).