Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Partially Identified Models”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 recordsLinked to original sources
PubMed2025

Uncertainty Modeling Outperforms Machine Learning for Microbiome Data Analysis.

Microbiome sequencing measures relative rather than absolute abundances, providing no direct information about total microbial load. Normalization methods attempt to compensate, but rely on strong, often untestable assumptions that can bias inference. Experimental measurements of load (e.g., qPCR, flow cytometry) offer a solution, but remain costly and uncommon. A recent high-profile study proposed that machine learning could bypass this limitation by predicting microbial load from sequencing data alone. To evaluate this claim, we assembled mutt, the largest public database of paired sequencing and load measurements, spanning 35 studies and over 15,000 samples. Using mutt, we show that published machine learning models fail to generalize: on average they perform worse than a naive baseline that always predicted the training set mean. These failures stem from covariate shift-limited shared taxa between studies, differences in community composition, and differences in preprocessing pipelines-that silently derail model inputs. In contrast, Bayesian partially identified models do not attempt to impute microbial load, but instead propagate scale uncertainty through downstream analyses. Across 30 benchmark datasets, Bayesian partially identified models consistently outperformed normalization and machine learning approaches, providing a principled and reproducible foundation for microbiome inference.

16S rRNA-seq↗
PubMed2026

Murine metabolic HFpEF is associated with altered mitochondrial substrate handling and S-nitrosylation remodeling.

Heart failure with preserved ejection fraction (HFpEF) is a heterogeneous condition with incompletely defined myocardial mechanisms. Here, using a two-hit murine model of cardiometabolic HFpEF induced by high-fat diet and endothelial nitric oxide synthase inhibition, we define a mitochondrial metabolic phenotype characterized by altered substrate handling, redox stress, and S-nitrosylation remodeling. While global proteomic changes were modest, metabolomic profiling revealed selective remodeling of tricarboxylic acid cycle intermediates, increased dicarboxylic acids, and altered redox-associated metabolites, consistent with mitochondrial metabolic and redox imbalance in this experimental setting. S-nitrosylation proteomics demonstrated a highly organized and bidirectional remodeling pattern affecting proteins involved in fatty acid/lipid metabolism, carbohydrate metabolism, mitochondrial energy metabolism, amino acid and organic acid metabolism, nucleotide/co-factor metabolism, and redox defense. Stable isotope tracing showed reduced glucose-derived and increased palmitate-derived acetyl-CoA in HFpEF, whereas Na-βHB reduced palmitate contribution and increased βHB-derived acetyl-CoA without restoring glucose contribution, indicating substrate redistribution and preserved ketone oxidation. Na-βHB supplementation increased oligomycin-sensitive respiration in freshly prepared left ventricular tissue, partially normalized selected TCA-cycle intermediates, reduced mitochondrial ROS and the NADH/NAD+ ratio, restored the GSH/GSSG ratio, and improved diastolic function without altering ejection fraction. Together, these findings define a redox-sensitive mitochondrial metabolic state in the HFD/l-NAME model and identify ketone supplementation as a partial metabolic rescue strategy in this context. At the same time, these findings highlight an important limitation of the murine HFD/l-NAME model, which should be interpreted as an experimental system for studying high-fat-induced cardiometabolic stress rather than as a metabolic equivalent of human HFpEF.

Animals↗
PubMedDate not supplied

Action of antiepileptic drugs against kainic acid-induced seizures and automatisms during ontogenesis in rats.

Kainic acid (KA 4-14 mg/kg) administered intraperitoneally (i.p.) produces automatisms (scratching until third postnatal week, "wet dog" shakes thereafter), and clonic and tonic-clonic seizures in rats aged 7, 12, 18, 25, and 90 days. Administration of carbamazepine (CBZ) i.p. (25 or 50 mg/kg), phenobarbital (PB 20-80 mg/kg), clonazepam (CZP 0.2 or 1 mg/kg), or valproate (VPA 200 mg/kg) influenced neither incidence nor latency of automatisms. Clonic seizures that are regularly observed after the third postnatal week in controls were either abolished or substantially suppressed by any of the aforementioned antiepileptic drugs (AEDs). Tonic-clonic seizures observed in the first 3 postnatal weeks were suppressed only by solvent [including propyleneglycol (PEG), ethanol, and water]; the effect of AEDs on tonic-clonic seizures was proconvulsant instead. The automatisms were most resistant to AED therapy. These results induce some doubts about the adequacy of the KA model for identifying AEDs effective against complex partial seizures, but forthcoming AEDs that suppress automatisms in the KA rat model might also be active against human complex partial seizures.

Animals↗
PubMed1991

A new method of estimating prevalence of injecting drug use in an urban population: results from a Scottish city.

Using modelling techniques derived from behavioural ecology, unnamed identifier data from a variety of partial samples of injecting drug users [IUDs] were used to estimate the prevalence of this population in Glasgow during 1989. The model yielded an estimate of 9424 [95% confidence interval +/- 2460] which represents a prevalence rate of 15 per 1000 population aged 15-55. The estimated male:female ratio was 2.64:1 and the modal IDU aged 20-24. These figures should facilitate assessment of the service requirements of this group and provide a basis from which the number of IDUs infected with HIV and the number likely to progress to AIDS can be determined. As the methodology required to facilitate reliable prevalence estimates utilizes relatively easy to obtain information it could be repeated in other urban centres where drug injecting is known to occur.

Adolescent↗
PubMed1978

Electro-optical studies on synthetic polyelectrolytes. IV. Electric polarizability and conformation of poly-N-methyl-2-vinylpyridinium bromide in aqueous solution.

Measurements of the relaxation time on aqueous solutions of the title polyelectrolyte as a function of the concentration and of the molecular weight show that its conformation at very high dilution can be better accounted for by a weakly bending rod or wormlike chain model, with persistence length ranging from 200 to 400 A. The analysis of the field strength dependence of the electric birefringence yields electric polarizability values which increase sharply with the dilution and are not significantly dependent upon the molecular weight. This has been tentatively interpreted on the basis of the theories derived by Oosawa and by Mandel and Van der Touw. The partially flexible model proposed by the latter authors allow to identify the electric polarizability calculated from electro-optical data to the specific dielectric increment measured in the high frequency range; both parameters appear to be molecular weight independent. The electric polarizability obtained from the Kerr effect would be originated in the induced dipoles caused by the delocalization of the bound counterions along rigid subunits whose length seems however to differ from the persistence length.

Birefringence↗
PubMed2026

Mechanisms by which carbamoylated high-density lipoprotein (C-HDL) promotes calcific aortic valve disease and exploration of potential targeted therapies.

Calcific aortic valve disease (CAVD) is a progressive fibrocalcific illness for which no effective pharmaceutical treatment exists. This study investigated whether carbamoylated high-density lipoprotein (C-HDL), a defective type of HDL that can develop during inflammation, contributes to CAVD progression and the involved molecular pathways. Male ApoE-/- mice were divided into three groups: CAVD model, cyanate-treated, and inhibitor, and analyzed after 12 weeks. C57BL/6 mice on a regular diet served as blank controls. Serum paraoxonase-1 (PON1), aortic valve calcification, cluster of differentiation 31 (CD31), phosphorylated nuclear factor kappa B p65 (p-p65), NOTCH receptor 1 (NOTCH1), and runt-related transcription factor 2 (RUNX2) were evaluated. In parallel, using RNA sequencing (RNA-seq), Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses, protein-protein interaction (PPI) network analysis, and quantitative real-time polymerase chain reaction. Cyanate treatment reduced serum PON1 levels, increased von Kossa-positive calcium deposition, and raised CD31, p-p65, NOTCH1, and RUNX2 levels compared with the model group, but Gly partially corrected these effects. Transcriptomic research identified 270 C-HDL-associated differentially expressed genes (DEGs) enriched in pathways associated with inflammatory signaling and NF-κB activity. Five potential hub genes (BIRC6, PIK3R1, ATM, IFIH1, and DDX58) were discovered and verified using qRT-PCR. These data show that C-HDL may accelerate CAVD by disrupting valve endothelial homeostasis and stimulating inflammatory signaling, and they identify potential molecular targets for future functional validation.

bioinformatics↗
PubMed2026

Spindle Assembly Checkpoint Competency Determines Sensitivity to KIF18A Inhibition in Small-Cell Lung Cancer.

BACKGROUND: Small-cell lung cancer (SCLC) is characterized by pervasive chromosomal instability (CIN) and remains largely refractory to targeted therapies. KIF18A, a motor protein that regulates chromosome alignment during mitosis, has emerged as a selective dependency in CIN-high tumors. Whether this dependency extends to SCLC, a prototypical CIN-high cancer, has not been established, and biomarkers predicting response to KIF18A inhibition, currently in clinical trials, are lacking. METHODS: We integrated analyses of patient tumor datasets, neuroendocrine (NE) and non- NE SCLC cell lines, and functional perturbation models to define the determinants of response to KIF18A inhibition. Chromosomal instability metrics, transcriptional programs, mitotic dynamics, and spindle assembly checkpoint (SAC) function were assessed using genomic profiling, live-cell imaging, genetic perturbation, and pharmacologic inhibition. RESULTS: KIF18A expression was elevated in SCLC tumors and correlated with CIN-associated transcriptional programs, proliferative markers, and NE status; however, these features did not predict sensitivity to KIF18A inhibition. Instead, response was determined by the functional integrity of the SAC. SAC-proficient SCLC cells underwent sustained mitotic arrest followed by apoptotic cell death upon KIF18A inhibition, whereas SAC-defective cells failed to maintain checkpoint activation and survived. Mechanistically, resistant cells exhibited impaired kinetochore recruitment of core SAC components, including MAD1 and BUBR1. Importantly, transient induction of acute CIN through MPS1 inhibition partially restored sensitivity to KIF18A inhibition in resistant models. CONCLUSIONS: This study provides the first mechanistic characterization of KIF18A dependency in SCLC, identifying SAC competency as the primary determinant of response. These findings establish a biologically informed framework for patient stratification and rational combination strategies. TRANSLATIONAL RELEVANCE: Small-cell lung cancer (SCLC) is an aggressive malignancy with few effective targeted therapies and marked chromosomal instability. KIF18A has emerged as a potential therapeutic target in genomically unstable cancers, but biomarkers predicting response to KIF18A inhibition are lacking. We demonstrate that sensitivity to KIF18A inhibition in SCLC is determined not by KIF18A expression, neuroendocrine subtype, or baseline chromosomal instability, but by the functional integrity of the spindle assembly checkpoint (SAC). SCLC cells with intact SAC signaling undergo sustained mitotic arrest and apoptosis upon KIF18A inhibition, whereas SAC-defective cells bypass checkpoint activation and survive aberrant mitosis. Notably, transient induction of acute chromosomal instability through MPS1 inhibition partially restores sensitivity in resistant models. Together, these findings identify mitotic checkpoint competency as a mechanistic determinant and candidate predictive biomarker for KIF18A-targeted therapies, providing a biologically informed framework for patient stratification and rational combination strategies relevant to ongoing KIF18A inhibitor clinical trials.

Journal Article↗
PubMed2026

Shared genetic architecture and therapeutic targets across paediatric immune-mediated diseases.

OBJECTIVES: Paediatric-onset immune-mediated inflammatory diseases (IMIDs), including juvenile idiopathic arthritis and related rheumatic diseases, remain genetically undercharacterised. We aimed to define shared and category-specific genetic architecture across paediatric IMIDs, compare signals with adult IMIDs, and identify therapeutic opportunities. METHODS: We analysed 24 paediatric IMIDs classified as autoimmune, polygenic-autoinflammatory, mixed-pattern, or allergic. Genome-wide association analyses included 18,086 cases and 131,019 controls of European ancestry. We estimated single nucleotide polymorphism (SNP)-based heritability, genetic correlations, and polygenic overlap; performed subset-based meta-analysis; and conducted functional annotation, gene prioritisation, pathway and protein network analyses, adult-IMID comparison, and drug-target prioritisation. RESULTS: SNP-based heritability ranged from 28.9% for allergic IMIDs to 61.9% for autoimmune IMIDs. Genetic correlation and polygenic modelling supported partial sharing across categories with category-specific components. Meta-analysis identified 39 genome-wide significant loci outside the Major Histocompatibility Complex (MHC) region, including 15 previously unreported loci; 19 loci were shared between categories. Gene-prioritisation and protein interaction analyses identified a core MHC-centred antigen-presentation network, with category-enriched modules involving complement, innate/barrier pathways, epithelial biology, and type 2 immunity. Enriched pathways included nuclear factor κB signalling, T helper 17 related pathways, Janus kinase-signal transducer and activator of transcription signalling, programmed cell death protein 1/programmed death‑ligand 1, cytotoxic T‑lymphocyte associated protein 4 regulation, and osteoclast differentiation, several of which are relevant to rheumatic diseases. Paediatric IMIDs shared broad polygenic architecture with adult IMIDs, whereas top-ranked genes converged strongly with adult rheumatic diseases. Priority Index analysis identified 178 high-scoring genes, including 43 approved or investigational IMID drug targets. CONCLUSIONS: Paediatric-onset IMIDs share core pathways with adult forms but exhibit distinct genetic architecture shaped by age-specific immune and neurodevelopmental biology. These findings provide a genomic framework for paediatric precision medicine, guiding classification, risk prediction, and therapeutic development.

Humans↗
PubMed1991

A mathematical model for radiation-induced myelopoiesis.

A model for damage, repair, killing, and repopulation of myelopoietic marrow is presented. Evaluation produces time and dose-rate profiles during and following any complex irradiation. Equations model variable dose rates, multiple exposures, different sources, and arbitrary intervals between treatments. If factors which dominate the control of biological processes can be demonstrated, an option is to set biological rate constants to experimentally determined values. Previously, knowledge did not permit identification of dominating biological processes and their temporal rates. But a unique feature of this study is that unspecified lesions for killing and injury of cells are evaluated from mortality data on the animal species of choice. "Unspecified" is used to indicate a condition of assumption-free modeling of molecular processes, whereby rate constants for cellular effects are simply computed directly from animal mortality data. Coefficients (estimated by maximum-likelihood methods for nonspecific processes) are compared with experimental values for specific processes. The model has many uses, including modeling of the myelopoietic potential as a function of time. Another option is to calculate the whole-body survival curve for cells that control myelopoiesis as a result of the treatment schedule. Also through simple extensions of the model, an extremely complex protocol can be identified with an equivalent prompt dose value--even for partial-body, fractionated exposures.

Animals↗
PubMed2024

A computational model for bacteriophage ϕX174 gene expression.

Bacteriophage ϕX174 has been widely used as a model organism to study fundamental processes in molecular biology. However, several aspects of ϕX174 gene regulation are not fully resolved. Here we construct a computational model for ϕX174 and use the model to study gene regulation during the phage infection cycle. We estimate the relative strengths of transcription regulatory elements (promoters and terminators) by fitting the model to transcriptomics data. We show that the specific arrangement of a promoter followed immediately by a terminator, which occurs naturally in the ϕX174 genome, poses a parameter identifiability problem for the model, since the activity of one element can be partially compensated for by the other. We also simulate ϕX174 gene expression with two additional, putative transcription regulatory elements that have been proposed in prior studies. We find that the activities of these putative elements are estimated to be weak, and that variation in ϕX174 transcript abundances can be adequately explained without them. Overall, our work demonstrates that ϕX174 gene regulation is well described by the canonical set of promoters and terminators widely used in the literature.

Gene Expression Regulation, Viral↗
PubMed1991

Residual plots for log odds ratio regression models.

This article describes graphical diagnostic methods for log odds ratio regression models. To study the effects of an additional covariate on log odds ratio regression analysis, three types of residual plots based on weighted least squares (WLS) are discussed: (i) added variable plot (partial regression plot), (ii) partial residual plot, and (iii) augmented partial residual plot. These plots provide diagnostic procedures for identifying heterogeneity of error variances, outliers, or nonlinearity of the model. They are especially useful for clarifying whether including a covariate as a linear term is appropriate, or whether quadratic or other nonlinear transformations are preferable. A well-known data set for case-control studies is analyzed to illustrate the residual plots.

Age Factors↗
PubMed2024

Elucidating the clinical and genetic spectrum of inositol polyphosphate phosphatase INPP4A-related neurodevelopmental disorder.

PURPOSE: Biallelic INPP4A variants have recently been associated with severe neurodevelopmental disease in single-case reports. Here, we expand and elucidate the clinical-genetic spectrum and provide a pathomechanistic explanation for genotype-phenotype correlations. METHODS: Clinical and genomic investigations of 30 individuals were undertaken alongside molecular and in silico modelling and translation reinitiation studies. RESULTS: We characterize a clinically variable disorder with cardinal features, including global developmental delay, severe-profound intellectual disability, microcephaly, limb weakness, cerebellar signs, and short stature. A more severe presentation associated with biallelic INPP4A variants downstream of exon 4 has additional features of (ponto)cerebellar hypoplasia, reduced cerebral volume, peripheral spasticity, contractures, intractable seizures, and cortical visual impairment. Our studies identify the likely pathomechanism of this genotype-phenotype correlation entailing translational reinitiation in exon 4 resulting in an N-terminal truncated INPP4A protein retaining partial functionality, associated with less severe disease. We also identified identical reinitiation site conservation in Inpp4a-/- mouse models displaying similar genotype-phenotype correlation. Additionally, we show fibroblasts from a single affected individual exhibit disrupted endocytic trafficking pathways, indicating the potential biological basis of the condition. CONCLUSION: Our studies comprehensively characterize INPP4A-related neurodevelopmental disorder and suggest genotype-specific clinical assessment guidelines. We propose that the potential mechanistic basis of observed genotype-phenotype correlations entails exon 4 translation reinitiation.

Humans↗
PubMed1992

Medium-term bioassays for carcinogens.

Medium-term bioassay systems in rats have been developed for rapid detection of carcinogenic agents and have been introduced for practical use. There are two major systems: N-nitrosodiethylamine and partial hepatectomy in liver and multiorgan models. The first model takes eight weeks; it consists of an initial intraperitoneal injection of N-nitrosodiethylamine, six weeks' administration of test chemical(s) beginning two weeks later, partial hepatectomy at week 3, and analysis of immunohistochemically identified glutathione S-transferase placental form-positive liver-cell foci. Using this model, the potency of carcinogenic agents can be predicted satisfactorily, at least for those of which the liver is a target organ. The multi-organ models include wide-spectrum organ initiation by N-methyl-N-nitrosourea or combined, short treatment with several carcinogens sequentially and subsequent administration of test chemical(s) for 12-20 weeks. Although these systems require a longer time, their accuracy as organotropism-independent testing systems has been validated by whole-body histological examination. These approaches promise precise, rapid detection of carcinogenic agents and should bridge the gap between established methods for screening in vitro and long-term carcinogenicity testing, with their inherent disadvantages. Other medium-term bioassay methods are discussed in addition to our two systems.

Animals↗
PubMed1977

Association of chromosome loss with centromere-adjacent mitotic recombination in a yeast disomic haploid.

Experiments designed to characterize the association between disomic chromosome loss and centromere-adjacent mitotic recombination were performed. Mitotic gene convertants were selected at two heteroallelic sites on the left arm of disomic chromosome III and tested for coincident chromosome loss. The principal results are: (1) Disomic chromosome loss is markedly enhanced (nearly 40-fold) over basal levels among mitotic gene convertants selected to arise close to the centromere; no such enhancement is observed among convertants selected to arise relatively far from the centromere. (2) Chromosome loss is primarily associated with proximal allele conversion at the centromere-adjacent site, and many of these convertants are reciprocally recombined in the adjacent proximal interval. (3) Partial aneuploid exceptions provisionally identified as carrying left arm telocentrics have been found. A testable model is proposed suggesting that centromere involvement in genetic recombination may precipitate segregational disfunction leading to mitotic chromosome loss.

Alleles↗
PubMed1992

Brugia malayi: antibody responses to larval antigens in infected and immunized jirds.

Vaccination with irradiated third stage Brugia malayi larvae (L3) has been reported to induce partial protective immunity to L3 challenge in jirds. The purpose of this study was to identify antigens that may be targets of protective immunity in this model. Jirds were immunized by s.c. injection of irradiated L3 and challenged either s.c. or i.p. Necropsy was performed 11 wk after challenge. Partial protection was achieved in s.c. challenged animals; worm recovery was only 41% of that observed in unvaccinated controls, and worms recovered from immunized animals were stunted. Worm recoveries in immunized animals that were challenged i.p. did not differ from those of unimmunized controls. Group differences in parasite antigen levels in sera collected 2-11 wk after larval challenge were consistent with parasitological findings obtained at necropsy. Antibody studies compared prechallenge sera from immunized animals to sera from infected (unimmunized) controls. Antibody responses to L3 surface antigens (assessed by IFA) were much stronger after immunization than after infection. Immunoblot studies showed preferential recognition of several L3 antigens (97, 54, 48, and 40 kDa) by antibodies in sera from immunized animals. Additional studies are needed to determine whether immunization with such preferentially recognized antigens can induce protection to larval challenge comparable to or better than that observed with live vaccines.

Animals↗
PubMed1992

The immune system and hypertension.

Primary or secondary activation of immune mechanisms has been implicated in the pathogenesis of many forms of hypertension. Changes in serum immunoglobulin levels, alterations in both humoral and cellular immune functions, and inherited abnormalities of the complement system have been identified in patients with essential hypertension. In addition, many models of spontaneous hypertension (such as the Okamoto and Lyon strains of hypertensive rats and the hypertensive New Zealand Black mouse) have identifiable abnormalities in immune function that are associated with their hypertensive disease. Other models (such as partial renal infarct hypertension, post-mineralocorticoid-salt hypertension, and hypertension induced by repeated injections of angiotensin II) also may have primary or secondary immunologic factors contributing to their etiology. Although there is a strong association between alterations in immune function and hypertension, the specific immunologic mechanisms that contribute to the pathogenesis of hypertension are not known. Therefore, further investigation will be necessary to elucidate these mechanisms.

Animals↗
PubMedDate not supplied

Acetylator phenotype-dependent and -independent expression of arylamine N-acetyltransferase isozymes in rapid and slow acetylator inbred rat liver.

Although mouse, hamster, and rabbit models of the human N-acetylation polymorphism have been identified and characterized, many investigations of arylamine toxicity and carcinogenicity are carried out in the rat, particularly the Fischer 344 (F-344) inbred rat. We partially characterized a new rat model of the N-acetylation polymorphism by determining expression of arylamine N-acetyltransferase activities in liver cytosols derived from adult male inbred F-344, WKY, and their F1 hybrid rat strains. Levels of N-acetyltransferase activity differed significantly between the strains for many arylamine substrates, with highest levels in F-344, lowest levels in WKY, and intermediate levels in F1 hybrids of these two parental strains. However, for some other arylamine substrates, levels of N-acetyltransferase activity did not differ significantly between the rat strains. Partial purification of rat liver cytosols from the three strains resulted in identification of two N-acetyltransferase isozymes. The levels of N-acetyltransferase activity of one isozyme differed significantly between strains analogous to the pattern observed in crude cytosol. In contrast, the levels of N-acetyltransferase activity of the second isozyme did not differ between the strains. Based upon these results, the F-344 inbred strain is designated a rapid acetylator phenotype, the WKY inbred strain is designated a slow acetylator phenotype, and F1 hybrids of the two parental strains are designated intermediate acetylator phenotype. The identification of acetylator phenotype-dependent and -independent hepatic N-acetyltransferase isozymes in the inbred rat mimics the biochemical basis for acetylator phenotype-dependent and -independent expressions of N-acetylation in humans and other mammalian species.

Acetylation↗
PubMed2026

Integrated Metabolomic and Transcriptomic Analysis Reveals Tissue-Specific Secondary Metabolic Differentiation and Indole Alkaloid Accumulation in Evodia rutaecarpa.

Evodia rutaecarpa is a valuable medicinal plant, yet its non-medicinal tissues remain largely underexplored. Here, we integrated ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS)-based widely targeted metabolomics and RNA sequencing (RNA-seq) transcriptomics to systematically profile the metabolic and transcriptional landscapes of roots, stems, leaves, and flowers of Evodia rutaecarpa (Juss.) Benth. Our aim was to characterize tissue-specific metabolic differentiation and its underlying transcriptional regulatory mechanisms. Metabolomic analysis, employing principal component analysis (PCA) and orthogonal partial least squares-discriminant analysis (OPLS-DA) with robust model parameters (R2Y > 0.9, Q2 > 0.5), identified 3090 differential metabolite features (variable importance in projection, VIP > 1.0; p < 0.05) across the four tissues, which exhibited distinct tissue-specific clustering patterns. Integrated Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis and weighted gene co-expression network analysis (WGCNA) revealed that roots specifically accumulated quinolone alkaloids and flavonoid glycosides, accompanied by the coordinated upregulation of genes involved in flavonoid and phenylpropanoid biosynthetic pathways. In contrast, stems, leaves, and flowers were enriched in indole alkaloids (evodiamine and rutaecarpine) and volatile oil precursors, with concurrent upregulation of genes involved in tryptophan metabolism and indole alkaloid biosynthesis (e.g., tryptophan decarboxylase, TDC; s N-methyltransferase, NMT). Notably, leaves and flowers displayed particularly high accumulation levels of these bioactive alkaloids, suggesting their potential as alternative sources for industrial and pharmaceutical applications. WGCNA further identified multiple transcription factors and structural gene modules tightly correlated with evodiamine accumulation, offering promising candidate regulators for future biosynthetic pathway engineering. Collectively, this multi-omics integration study systematically elucidates the tissue-partitioned secondary metabolism of Evodia rutaecarpa (Juss.) Benth. and provides a solid scientific foundation for full-plant resource utilization, targeted development of non-medicinal tissues, and future metabolic engineering of indole alkaloid production.

Evodia rutaecarpa↗