• HOME
  • NEWS
  • EXPLORE
    • CAREER
      • Companies
      • Jobs
    • EVENTS
    • iGEM
      • News
      • Team
    • PHOTOS
    • VIDEO
    • WIKI
  • BLOG
  • COMMUNITY
    • FACEBOOK
    • INSTAGRAM
    • TWITTER
Thursday, November 19, 2026
BIOENGINEER.ORG
No Result
View All Result
  • Login
  • HOME
  • NEWS
  • EXPLORE
    • CAREER
      • Companies
      • Jobs
        • Lecturer
        • PhD Studentship
        • Postdoc
        • Research Assistant
    • EVENTS
    • iGEM
      • News
      • Team
    • PHOTOS
    • VIDEO
    • WIKI
  • BLOG
  • COMMUNITY
    • FACEBOOK
    • INSTAGRAM
    • TWITTER
  • HOME
  • NEWS
  • EXPLORE
    • CAREER
      • Companies
      • Jobs
        • Lecturer
        • PhD Studentship
        • Postdoc
        • Research Assistant
    • EVENTS
    • iGEM
      • News
      • Team
    • PHOTOS
    • VIDEO
    • WIKI
  • BLOG
  • COMMUNITY
    • FACEBOOK
    • INSTAGRAM
    • TWITTER
No Result
View All Result
Bioengineer.org
No Result
View All Result
Home NEWS Science News Biology

Computational Analysis Identifies Overlapping Molecular Targets Between Aspartame and Pulmonary Hypertension

by
November 11, 2026
in Biology, Technology
Reading Time: 6 mins read
0
Computational Analysis Identifies Overlapping Molecular Targets Between Aspartame and Pulmonary Hypertension

Computational Analysis Identifies Overlapping Molecular Targets Between Aspartame and Pulmonary Hypertension

Share on FacebookShare on TwitterShare on LinkedinShare on RedditShare on Telegram

Computational researchers have uncovered preliminary molecular intersections between the biological target profiles of the nonnutritive sweetener aspartame and gene expression signatures characteristic of pulmonary arterial hypertension. Using an exploratory framework that combined network toxicology, weighted co-expression network analysis, and machine learning algorithms, investigators evaluated whether molecular pathways linked to the artificial sweetener share common biological ground with the progressive vascular disease. Across public bulk lung transcriptomic datasets, the computational pipeline detected twenty-one overlapping candidate genes enriched in regulatory pathways governing vascular homeostasis, including the renin-angiotensin system and calcium-related cellular processes. A five-gene predictive panel prioritizing CAPN1, ERAP2, ITK, PTGS1, and KDR was subsequently identified through machine learning model combinations, demonstrating high classification accuracy within discovery lung tissue samples. However, the study authors emphasize that these computational observations represent hypothesis-generating findings rather than established clinical associations, as the source datasets lacked dietary exposure records, circulating metabolite quantification, and direct experimental validation in living physiological systems.

Pulmonary arterial hypertension represents an uncommon, chronic, and progressive cardiovascular disorder defined by persistently elevated pulmonary arterial pressure, progressive vascular remodeling, sustained vasoconstriction, perivascular inflammation, and eventual right ventricular hypertrophy. Global epidemiological analyses from the 2021 Global Burden of Disease study estimate an age-standardized prevalence of approximately 2.28 cases per 100,000 population, accompanied by an annual mortality rate of 0.27 deaths per 100,000 individuals worldwide. Current standard pharmacotherapies target the nitric oxide, prostacyclin, and endothelin pathways to slow functional decline, yet clinical management remains challenging, with reported perioperative mortality reaching up to eighteen percent in elective noncardiac operations and fifty percent during emergency interventions. Concurrently, public health authorities have maintained acceptable daily intake benchmarks for aspartame, set at forty milligrams per kilogram of body weight per day by international food safety bodies and fifty milligrams per kilogram by the United States Food and Drug Administration. Following oral ingestion, intact aspartame undergoes rapid enzymatic hydrolysis in the gastrointestinal tract into phenylalanine, aspartic acid, and methanol, meaning systemic vascular exposure predominantly reflects circulating metabolites rather than the parent sweetener.

To investigate potential molecular intersections between aspartame-related target networks and pulmonary arterial hypertension, investigators retrieved and processed four human microarray transcriptomic datasets archived within the NCBI Gene Expression Omnibus repository. The discovery cohort combined bulk lung tissue profiles from datasets GSE113439 and GSE117261, while external validation cohorts incorporated peripheral blood mononuclear cell profiles derived from datasets GSE131793 and GSE33463. Inclusion criteria across all datasets required confirmed clinical diagnoses of pulmonary arterial hypertension alongside non-hypertensive reference controls. To account for systematic platform variations and technical artifacts between separate profiling experiments, investigators applied ComBat empirical Bayes batch correction algorithms implemented within the sva software environment, subsequently confirming reduced dataset-associated separation using principal component analysis. For chemical target identification, researchers retrieved the molecular structure of aspartame from the PubChem chemical database and gathered predicted human biological targets across PubChem, ChEMBL, and SuperPred repository queries. After programmatic removal of duplicate target accessions, the curated query yielded one hundred and sixty-three unique candidate human protein targets associated with the compound for subsequent integrative network analysis.

Differential expression profiling within the discovery lung tissue cohort was conducted using the limma analytical package, applying statistical significance thresholds of Benjamini-Hochberg adjusted P-values below 0.05 and absolute log2 fold change values exceeding 0.585, which corresponds to a 1.5-fold difference in expression. This filtering strategy identified one hundred and sixty-six differentially expressed genes across the heterogeneous bulk lung tissue samples. In parallel, weighted gene co-expression network analysis was applied to discovery samples following hierarchical sample clustering for quality assurance. Selecting a soft-thresholding power of beta equals six established scale-free topological architecture with a model fit coefficient exceeding 0.85, resolving ten distinct gene co-expression modules through dynamic tree cutting. Among these, the turquoise module demonstrated the strongest statistically significant positive correlation with pulmonary arterial hypertension status, encompassing 1,703 co-expressed genes. Merging these module constituents with the differentially expressed gene set yielded a comprehensive disease-associated baseline of 1,839 candidate genes, which shared thirty overlapping transcripts between the two analytical derivation methods.

Cross-referencing the 163 predicted aspartame targets against the 1,839 pulmonary arterial hypertension-associated candidate genes revealed an intersection of twenty-one overlapping genes, a convergence deemed statistically significant by hypergeometric distribution testing against the background of 23,306 detected human genes. Functional annotation via Gene Ontology enrichment analyses demonstrated that these shared molecular targets were prominently clustered within biological processes governing receptor and peptide catabolism, nitric oxide-mediated cellular signaling, and calcium homeostasis, alongside cellular localizations comprising lysosomal and endocytic membranes. Pathway enrichment interrogation using the Kyoto Encyclopedia of Genes and Genomes database identified significant overrepresentation of the renin-angiotensin system, a central hormonal signaling cascade regulating blood pressure and vascular tone. Within vascular biology, the counter-regulatory balance between the angiotensin-converting enzyme axis and the protective angiotensin-converting enzyme 2 pathway exerts substantial influence over endothelium-dependent vasorelaxation, cardiac remodeling, and inflammatory suppression, thereby linking the computationally prioritized target collection to mechanisms established in pulmonary vascular remodeling.

To prioritize the most predictive biomarkers among the twenty-one overlapping candidates, researchers deployed a multimodel machine learning framework evaluating twelve distinct statistical and machine learning algorithms, including lasso regression, ridge regression, elastic net, support vector machines, random forests, and gradient boosting machines. Systematic combination of these algorithms yielded one hundred and thirteen model permutations optimized through five-fold internal cross-validation. An ensemble combining random forest and extreme gradient boosting architectures was selected as the top-performing classifier, retaining fifteen informative input features. Model interpretability was evaluated using SHapley Additive exPlanations, which ranked CAPN1, ERAP2, ITK, PTGS1, and KDR as the five most influential predictive features, with CAPN1 exhibiting the highest mean absolute SHAP value of 0.120. Directional SHAP analyses indicated that elevated expression of CAPN1 and KDR generally contributed to lower predicted disease probability, whereas increased expression of ERAP2, ITK, and PTGS1 increased predicted probability, with dependence plots revealing nonlinear interaction thresholds near normalized expression values of 7.5.

Evaluation of the prioritized five-gene panel in bulk lung tissue demonstrated strong diagnostic classification, yielding an area under the receiver operating characteristic curve of 0.972 with a ninety-five percent confidence interval spanning from 0.935 to 1.000. When tested independently within external peripheral blood mononuclear cell validation cohorts, the classifier achieved an area under the curve of 0.660 with a ninety-five percent confidence interval of 0.534 to 0.787. Assessing CAPN1 in isolation yielded an area under the curve of 0.823 in discovery lung tissue and 0.721 in external blood samples. Decision curve analysis indicated positive estimated clinical net benefit across threshold probabilities ranging between 0.20 and 0.75, though external calibration shifted between cohorts. Notably, ERAP2 and ITK fulfilled formal criteria for differential expression in bulk lung tissue, whereas CAPN1, PTGS1, and KDR did not meet differential expression thresholds, illustrating that machine learning feature prioritization captured network-level multivariate patterns that diverged from conventional univariate mean expression differences.

To examine direct physical interactions between aspartame and the prioritized candidate proteins, investigators conducted structural molecular docking simulations using the CB-Dock2 computational platform, which integrates cavity detection with the AutoDock Vina scoring engine. Three-dimensional crystal structures for CAPN1, ERAP2, ITK, PTGS1, and KDR were docked against the intact aspartame molecular structure, producing negative binding affinity scores ranging between minus 8.0 and minus 5.7 kilocalories per mole, with the strongest predicted affinity observed for the aminopeptidase ERAP2. Visualized binding conformations revealed predicted hydrogen-bond contacts within identified catalytic pockets. Comparative reference docking against the calcium-sensing receptor yielded scores of minus 9.9 kilocalories per mole for the allosteric modulator cinacalcet and minus 6.3 kilocalories per mole for the aspartame metabolite phenylalanine. Investigators noted that while these docking scores supply theoretical structural feasibility, docking calculations across distinct protein cavities cannot establish physiological target engagement, particularly given the near-complete presystemic breakdown of dietary aspartame prior to reaching pulmonary vasculature.

To resolve cell type-specific expression dynamics, researchers mined single-cell RNA sequencing data from twelve human lung donors within dataset GSE293580, encompassing four healthy control donors, four patients with idiopathic pulmonary arterial hypertension, and four with systemic sclerosis-associated pulmonary arterial hypertension. Proportional cell distribution analyses revealed reduced hematopoietic-cell and neutrophil fractions in diseased lungs alongside expanded populations of fibroblasts, mast cells, and vascular smooth muscle cells. Single-cell mapping demonstrated that CAPN1, KDR, PTGS1, and ERAP2 were preferentially localized within annotated hematopoietic populations, whereas ITK expression predominated in natural killer and T cell subsets. Within hematopoietic cells, CAPN1 and CAPN2 expression was lower in hypertensive samples, while PLCG1 and CACNA1C showed higher expression, although composite calcium-signaling scores showed no significant difference. Simulating virtual CAPN1 knockout in healthy control hematopoietic cells using scTenifoldKnk tensor decomposition nominated VNN2 and MLX as significantly perturbed network effectors, suggesting downstream links to leukocyte transendothelial migration, oxidative stress responses, and cellular energy metabolism.

Despite the identification of candidate genetic overlaps, substantial biological and methodological limitations caution against drawing causal conclusions or clinical inferences regarding dietary aspartame consumption and pulmonary vascular risk. Public transcriptomic repositories lacked individual nutritional records, dietary exposure histories, and plasma metabolite measurements, leaving the actual physiological relevance of parent aspartame exposure unverified. Furthermore, the modest discrimination of the five-gene classifier in external blood samples, combined with shifted calibration curves across patient cohorts, underscores that these retrospective computational models cannot currently guide clinical diagnostics or surgical risk stratification. The authors explicitly stress that static docking calculations, single-cell associations, and virtual gene perturbation networks remain exploratory models that require rigorous functional validation. Future research must deploy controlled exposure experiments evaluating whether aspartame metabolites such as phenylalanine influence pulmonary arterial remodeling through calcium-sensing receptor pathways and downstream calpain enzymatic activation, alongside prospective cohort studies with verified dietary monitoring before clinical significance can be established.

Subject of Research: Computational prioritization of candidate molecular intersections between aspartame targets and pulmonary arterial hypertension

Article Title: Exploring candidate molecular links between aspartame and pulmonary arterial hypertension through network toxicology and transcriptomics

Article References: Zhu, D., & Chen, X. (2026). Exploring candidate molecular links between aspartame and pulmonary arterial hypertension through network toxicology and transcriptomics. Frontiers in Bioinformatics, 6, Article 2008294. https://doi.org/10.3389/fbinf.2026.2008294

Image Credits: AI Generated

DOI: 10.3389/fbinf.2026.2008294

Keywords: aspartame, pulmonary arterial hypertension, network toxicology, transcriptomics, machine learning, WGCNA, CAPN1, ERAP2, ITK, PTGS1, KDR, single-cell RNA sequencing, molecular docking, renin-angiotensin system, calcium signaling, scTenifoldKnk, VNN2, MLX

News Source: Gregory Coleman. (November 11, 2026). Computational Analysis Identifies Overlapping Molecular Targets Between Aspartame and Pulmonary Hypertension. Scienmag.

Tags: aspartameCalcium signalingCAPN1ERAP2ITKKDRMachine LearningMLXmolecular dockingNetwork toxicologyPTGS1Pulmonary arterial hypertensionrenin-angiotensin systemscTenifoldKnksingle-cell RNA sequencingTranscriptomicsVNN2WGCNA
Share12Tweet7Share2ShareShareShare1

Related Posts

Ultrafast 4D Ultrasound Angiography Maps Deep Primate Cerebral Blood Flow Through Intact Skulls

Ultrafast 4D Ultrasound Angiography Maps Deep Primate Cerebral Blood Flow Through Intact Skulls

November 19, 2026
Host Genetics Shapes Caecal Microbiota Composition in White Leghorn Laying Hens

Host Genetics Shapes Caecal Microbiota Composition in White Leghorn Laying Hens

November 19, 2026

Native Sunflower Rhizobacterium Suppresses Root Rot and Triggers Host Defense Genes

November 19, 2026

Portable Molecular Laboratory Achieves Rapid Sickle Cell Genotyping in Community Trial

November 19, 2026

POPULAR NEWS

  • Neuropathic Pain Affects Over Forty Percent of Patients With Rare CANVAS Disorder

    Neuropathic Pain Affects Over Forty Percent of Patients With Rare CANVAS Disorder

    29 shares
    Share 12 Tweet 7
  • Novel Phenylthiazolylquinoline Derivatives Show Potent In Vitro Cancer Cell Growth Inhibition

    29 shares
    Share 12 Tweet 7
  • Review Examines Central Nervous System Risks From Low-Frequency Sound and Infrasound

    29 shares
    Share 12 Tweet 7
  • Surgery Fails to Outperform Conservative Care for Pain in Fibrous Dysplasia

    29 shares
    Share 12 Tweet 7

About

We bring you the latest biotechnology news from best research centers and universities around the world. Check our website.

Follow us

Recent News

Curved M1 Segment Linked to Higher Temporary Clipping Odds in Aneurysm Surgery

Ultrafast 4D Ultrasound Angiography Maps Deep Primate Cerebral Blood Flow Through Intact Skulls

Survey Links AI Training and Professional Title to Anesthesia Nurse AI Literacy

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 89 other subscribers
  • Contact Us

Bioengineer.org © Copyright 2023 All Rights Reserved.

Welcome Back!

Login to your account below

Forgotten Password?

Retrieve your password

Please enter your username or email address to reset your password.

Log In
No Result
View All Result
  • Homepages
    • Home Page 1
    • Home Page 2
  • News
  • National
  • Business
  • Health
  • Lifestyle
  • Science

Bioengineer.org © Copyright 2023 All Rights Reserved.