Skip to main content
Yearly Report

Respiratory Research Analysis

2024
10 papers selected
852 analyzed

Respiratory science in 2025 was defined by host-directed mechanisms, platform-ready biologics, and organelle-to-tissue–level mechanistic clarity. A protein language model generated de novo paired heavy/light-chain antibodies across multiple respiratory pathogens, inaugurating a rapid-response biologics paradigm. Two convergent lines of work reshaped antiviral strategy: discovery of MFSD6 as the bona fide EV-D68 receptor enabling decoy and receptor-blocking approaches, and a CRISPRa-guided identi

Summary

Respiratory science in 2025 was defined by host-directed mechanisms, platform-ready biologics, and organelle-to-tissue–level mechanistic clarity. A protein language model generated de novo paired heavy/light-chain antibodies across multiple respiratory pathogens, inaugurating a rapid-response biologics paradigm. Two convergent lines of work reshaped antiviral strategy: discovery of MFSD6 as the bona fide EV-D68 receptor enabling decoy and receptor-blocking approaches, and a CRISPRa-guided identification of P-selectin as a modifiable vascular adhesion axis with in vivo viral clearability. Foundational structural biology mapped the in-cell architecture of the mitochondrial respiratory chain and, together with an oxygen–mitochondrial citrate export axis, repositioned organelle metabolism as a driver of airway fate. An evolution-aware pipeline preemptively redesigned a clinical antibody to retain breadth, while RNA-level evolution via variant TRS/sgRNAs reframed interferon evasion and surveillance. Manufacturing quality emerged as biology: minimizing defective interfering particles expanded LAIV mucosal breadth. Early-life immunity was mechanistically linked to asthma risk via maternal allergy and neonatal RSV through FcRn/FcγR-mediated priming, sharpening prevention windows.

Selected Articles

1. Generation of antigen-specific paired-chain antibodies using large language models.

Cell · 2025PMID: 41192421

A protein language model generated de novo, human paired heavy/light-chain antibodies that bind antigens from SARS-CoV-2, H5N1, and RSV-A, demonstrating a template-free, cross-pathogen discovery platform for rapid biologics.

Impact: Establishes a generalizable, rapid-response route to antibody discovery that can compress timelines for countermeasures against emerging respiratory threats.

Clinical Implications: If in vivo neutralization, developability, and safety are validated, this approach could accelerate therapeutic and prophylactic antibody pipelines and improve outbreak readiness.

Key Findings

  • Language model generated paired VH/VL antibodies de novo from sequence alone.
  • Binding validated across SARS-CoV-2, H5N1, and RSV-A.
  • Produced novel and diverse sequences without structural templates.

Methodological Strengths

  • Template-free generative modeling of paired antibody chains with experimental binding validation.
  • Cross-pathogen demonstration spanning multiple high-priority respiratory threats.

Limitations

  • In vivo efficacy, developability, and safety were not yet established.
  • Potential biases from training data may influence sequence space exploration.

Future Directions: Validate neutralization breadth and pharmacology in vivo, integrate liability prediction and manufacturability filters, and operationalize rapid-response design-to-test pipelines.

Abstract not available from provided dataset; see summary.

2. MFSD6 is an entry receptor for enterovirus D68.

Nature · 2025PMID: 40132641

MFSD6 was established as the cellular entry receptor for EV-D68, providing a mechanistic basis for host tropism and enabling receptor-blocking and decoy strategies to prevent infection and severe outcomes including AFM.

Impact: Orthogonal receptor validation consolidates a tractable, host-directed intervention point against a pediatric-relevant respiratory neurotropic pathogen.

Clinical Implications: Enables development of receptor-blocking antibodies/small molecules and engineered decoys; supports tissue expression–based risk stratification.

Key Findings

  • MFSD6 is the bona fide entry receptor for EV-D68.
  • Mechanistic basis for host cell attachment and tropism.
  • Clear path to receptor-targeted therapies and decoys.

Methodological Strengths

  • Orthogonal validation across genetic, biochemical, and cell-entry assays.
  • Mechanistic mapping of receptor domains mediating viral recognition.

Limitations

  • Human in vivo efficacy and safety of receptor-targeted interventions remain to be tested.
  • Potential clade-specific differences require cross-lineage validation.

Future Directions: Advance MFSD6 decoys and blockers into pediatric-focused preclinical and early clinical studies; integrate receptor expression maps into surveillance and AFM risk modeling.

Abstract not available from provided dataset; see summary.

3. In-cell architecture of the mitochondrial respiratory chain.

Science · 2025PMID: 40112058

Using in-cell cryo-electron tomography, the study visualized native respiratory complexes and supercomplex assemblies in intact cells, linking spatial organization to electron transfer and proton pumping efficiency in vivo.

Impact: Delivers native-context structural biology that underpins bioenergetics and respiratory pathophysiology, enabling organelle-aware biomarkers and interventions.

Clinical Implications: Supports hypotheses for mitochondrial biomarkers and therapies that modulate supercomplex organization in respiratory diseases.

Key Findings

  • In situ visualization of respiratory complexes and supercomplexes in intact cells.
  • Structural basis connecting organization to electron transfer and proton pumping.
  • Foundation for linking mitochondrial architecture to disease phenotypes.

Methodological Strengths

  • State-of-the-art in-cell cryo-ET revealing native macromolecular organization.
  • Correlation of structural organization with functional bioenergetic readouts.

Limitations

  • Primarily structural and hypothesis-generating without direct clinical outcomes.
  • Generalizability across cell types and disease states requires further study.

Future Directions: Map disease- and hypoxia-induced remodeling of supercomplexes, develop drugs that modulate complex assembly, and validate organelle-level biomarkers in patient cohorts.

Abstract not available from provided dataset; see summary.

4. Live attenuated influenza vaccine with low proportions of defective interfering particles elicits robust immunogenicity and cross-protection.

Nature communications · 2025PMID: 41173917

Mouse studies showed that low-DIP LAIV induces stronger mucosal/systemic immunity and complete cross-protection against lethal H3N2/H1N1 challenges compared with high-DIP LAIV, elevating manufacturing control as an immunogenic determinant.

Impact: Links process engineering to immunogenic breadth, providing a generalizable lever for improving mucosal vaccines and preparedness.

Clinical Implications: Human validation of DIP minimization could enhance LAIV effectiveness and inform quality attributes for lot release in next-generation mucosal vaccines.

Key Findings

  • Low-DIP LAIV elicited stronger mucosal and systemic immunity than high-DIP LAIV.
  • Complete cross-protection against multiple lethal influenza challenges.
  • Enhanced antigen presentation and mucosal cellular responses under low DIP.

Methodological Strengths

  • Controlled manipulation and quantification of DIPs across vaccine lots.
  • Robust challenge models with multi-strain, lethal endpoints.

Limitations

  • Preclinical (murine) findings require human confirmation.
  • Manufacturing scalability and standardized DIP assays need establishment.

Future Directions: Develop validated DIP assays for human lots, test low-DIP LAIV in phase 2/3 trials, and generalize DIP control principles to other mucosal vaccine platforms.

Abstract not available from provided dataset; see summary.

5. The oxygen level in air directs airway epithelial cell differentiation by controlling mitochondrial citrate export.

Science Advances · 2025PMID: 39854459

Ambient oxygen directs airway epithelial differentiation by regulating mitochondrial citrate export, positioning citrate export as a metabolic control point linking oxygen tension to epithelial fate decisions.

Impact: Defines an oxygen–metabolism–differentiation axis that bridges organelle bioenergetics with airway biology and regenerative strategies.

Clinical Implications: Motivates optimization of oxygen tension and citrate/acetyl-CoA metabolism in airway organoids, and exploration of citrate-export modulation to tune epithelial composition.

Key Findings

  • Ambient oxygen levels steer airway epithelial differentiation.
  • Mitochondrial citrate export links oxygen to fate decisions.
  • Repositions oxygen as a metabolic/developmental cue in airway biology.

Methodological Strengths

  • Mechanistic dissection with metabolic flux manipulation and organoid models.
  • Convergent evidence across oxygen tension, mitochondrial transport, and fate mapping.

Limitations

  • Translational biomarker-readouts need validation in human airway tissues in vivo.
  • Potential context specificity across airway regions and disease states.

Future Directions: Engineer metabolic microenvironments in airway models, identify druggable nodes in citrate export, and test organelle-aware therapies for chronic airway disease.

Abstract not available from provided dataset; see summary.

6. Maternal allergy and neonatal RSV infection synergize via FcR-mediated allergen uptake to promote the development of asthma in early life.

Science immunology · 2025PMID: 41313755

Registry-linked analyses and neonatal models show that maternal allergen sensitization combined with neonatal RSV-like infection amplifies FcRn/FcγR-mediated allergen uptake, drives cDC2 maturation and Th2 programming, and increases subsequent asthma risk.

Impact: Mechanistically connects perinatal immune programming to childhood asthma and identifies modifiable timing and targets for prevention.

Clinical Implications: Supports risk stratification and maternal/infant-targeted interventions to interrupt FcRn/FcγR-mediated priming around RSV exposure.

Key Findings

  • Higher later asthma risk among infants with RSV bronchiolitis born to allergic/asthmatic parents.
  • Neonatal viral infection upregulates Fc receptors and matures cDC2, priming Th2 responses.
  • Maternal allergen-specific IgG via FcRn enhances FcγR-mediated allergen uptake and Th2 programming.

Methodological Strengths

  • Registry-linked human data integrated with mechanistic neonatal models.
  • Causal pathway dissection implicating FcRn/FcγR and dendritic cell programming.

Limitations

  • Translational interventions targeting Fc pathways in humans remain to be tested prospectively.
  • Population heterogeneity and environmental co-exposures may modulate effect sizes.

Future Directions: Design maternal/infant trials timing RSV season to interrupt Fc-mediated priming; identify biomarkers for early-life risk stratification and targeted prevention.

Abstract not available from provided dataset; see summary.

7. A deep generative model for deciphering cellular dynamics and in silico drug discovery in complex diseases.

Nature biomedical engineering · 2025PMID: 40542107

UNAGI models time-series single-cell transcriptomic dynamics to prioritize drug candidates; in pulmonary fibrosis, it predicted repurposable agents and validated nifedipine’s anti-fibrotic effects in human precision-cut lung slices with proteomic support.

Impact: Bridges single-cell disease trajectories with actionable drug prioritization and ex vivo validation, accelerating therapeutic discovery in respiratory fibrosis.

Clinical Implications: Offers a preclinical prioritization platform for repurposing and early-phase trials in pulmonary fibrosis; informs candidate and biomarker selection.

Key Findings

  • Captured time-resolved single-cell disease trajectories and improved drug perturbation modeling.
  • Predicted antifibrotic candidates; nifedipine validated ex vivo in human lung tissue.
  • Proteomic data corroborated inferred dynamics; framework generalized to other diseases.

Methodological Strengths

  • Integrative generative modeling with external ex vivo validation in human tissue.
  • Mechanistically interpretable trajectories supporting drug mechanism hypotheses.

Limitations

  • Prospective, multi-center clinical validation is needed to confirm translational impact.
  • Model performance may vary with dataset quality and sampling frequency.

Future Directions: Prospective deployment with biomarker-linked early-phase trials; expand to other respiratory diseases and integrate multi-omic measurements.

Abstract not available from provided dataset; see summary.

8. P selectin promotes SARS-CoV-2 interactions with platelets and the endothelium.

The Journal of clinical investigation · 2025PMID: 41243963

A CRISPRa screen identified P-selectin as a host factor that enhances spike-dependent binding and vascular homing/platelet aggregation; blocking P-selectin interactions cleared vascular-associated pulmonary virus in vivo.

Impact: Reveals a modifiable vascular adhesion axis with in vivo support, enabling host-directed antiviral adjuncts for severe coronavirus disease.

Clinical Implications: Supports development of P-selectin–targeted strategies to mitigate vascular sequestration and thromboinflammation in respiratory viral disease.

Key Findings

  • P-selectin emerged from CRISPRa as a validated suppressor of SARS-CoV-2 infection.
  • P-selectin increases spike-dependent binding and mediates vascular homing and platelet aggregation.
  • Blocking P-selectin interactions cleared pulmonary vascular virus in vivo.

Methodological Strengths

  • Genome-scale CRISPRa screening coupled with mechanistic validation.
  • In vivo functional blockade demonstrating therapeutic clearability.

Limitations

  • Virus–host interactions and clearance dynamics may vary across variants and species.
  • Clinical translation requires evaluation of safety, timing, and patient selection.

Future Directions: Test P-selectin inhibitors or blocking biologics in respiratory viral models and early-phase trials; integrate vascular biomarkers into risk stratification.

Abstract not available from provided dataset; see summary.

9. Preemptive optimization of a clinical antibody for broad neutralization of SARS-CoV-2 variants and robustness against viral escape.

Science Advances · 2025PMID: 40153503

An integrated pipeline of deep mutational scanning, structure-guided design, and machine learning redesigned a clinical antibody to restore and broaden neutralization across current and prospective escape variants while avoiding new vulnerabilities.

Impact: Provides a reusable, evolution-aware blueprint to future-proof monoclonal antibodies against rapidly evolving respiratory viruses.

Clinical Implications: Supports periodic computational updates of clinical antibodies to preserve prophylactic and therapeutic options, especially for immunocompromised patients.

Key Findings

  • DMS mapped vulnerability hotspots in the parental antibody.
  • Redesign improved potency and breadth across diverse variants.
  • No new susceptibility hotspots emerged in redesigned antibody by DMS.

Methodological Strengths

  • Closed-loop design integrating DMS constraints, structural modeling, and ML.
  • Prospective evaluation across current and hypothetical escape variants.

Limitations

  • Clinical efficacy against real-world variant waves remains to be confirmed.
  • Manufacturability and immunogenicity of redesigned antibodies require assessment.

Future Directions: Institutionalize periodic antibody redesign cycles tied to surveillance; extend to other respiratory pathogens and polyclonal modalities.

Abstract not available from provided dataset; see summary.

10. Emergence of SARS-CoV-2 subgenomic RNAs that enhance viral fitness and immune evasion.

PLoS Biology · 2025PMID: 39836705

Global analyses and functional experiments reveal convergently evolved TRSs that generate novel sgRNAs, including a truncated N sgRNA antagonizing type I interferon and increasing viral fitness, highlighting RNA-level evolution beyond amino acid changes.

Impact: Exposes an RNA-level evolutionary mechanism shaping interferon evasion and fitness, advocating TRS/sgRNA-aware surveillance and therapeutics.

Clinical Implications: Supports integrating TRS/sgRNA features into variant risk assessment and exploring antivirals targeting TRS-dependent transcription or sgRNA functions.

Key Findings

  • Convergent emergence of novel TRSs upstream of structural genes.
  • Truncated N sgRNA antagonizes type I interferon and confers fitness advantages.
  • Demonstrates functional RNA-level evolution beyond protein coding changes.

Methodological Strengths

  • Integration of global genomic surveillance with mechanistic validation.
  • Dissection of sgRNA functions and interferon antagonism in vitro and in models.

Limitations

  • Clinical correlates of TRS/sgRNA features require prospective validation.
  • Scope centered on SARS-CoV-2; generalization to other RNA viruses needs study.

Future Directions: Embed TRS/sgRNA metrics into variant triage and explore inhibitors of transcriptional regulatory switches or pathogenic sgRNAs.

Abstract not available from provided dataset; see summary.