: Large-scale analysis of cutaneous microbiome dynamics in dermatological diseases using multi-omics integration

Project Abstract

Large-scale analysis of cutaneous microbiome dynamics across common dermatological diseases reveals robust, disease-specific microbial signatures and host–microbiome interaction networks that evolve with disease progression and treatment. This study integrates multi-omics data from a diverse cohort of patients with atopic dermatitis, psoriasis, acne, seborrheic dermatitis, and rosacea, alongside healthy controls, to characterize temporal microbiome shifts, functional potential, and host inflammatory responses. Skin swab and biopsy samples were collected at baseline, mid-treatment, and post-treatment intervals, enabling longitudinal assessment. 16S rRNA gene sequencing, shotgun metagenomics, metatranscriptomics, and metabolomics were combined with host transcriptomics and proteomics to capture compositional changes, virulence factor abundance, microbial gene expression, metabolite flux, and host immune pathways. Advanced bioinformatic pipelines employed include strain-resolved metagenomics, network-based co-occurrence and interaction modeling, and causal inference approaches to disentangle microbe–host and microbe–microbe relationships. We further integrated clinical metadata, including disease severity scores, treatment regimens, prior antibiotic exposure, and environmental factors, to model determinants of microbiome dynamics using mixed-effects and machine learning frameworks. Key findings demonstrate that disease states are associated with distinct dysbiosis patterns, such as reduced microbial diversity and enrichment of Staphylococcus aureus in atopic dermatitis, Propionibacterium acnes in acne, and Malassezia-dominated signatures in seborrheic dermatitis, with psoriasis showing a unique bacterial- and fungal-communities shift coupled to keratinocyte inflammatory signaling. Functional analyses reveal enrichment of virulence-associated pathways, biofilm formation capacity, and lipid metabolism genes in disease-associated microbiomes, while healthy skin exhibits gene expression profiles indicative of barrier function maintenance and protective host antimicrobial responses. Metabolomic data identify disease-specific signatures, including altered short-chain fatty acid, bile acid, and lipid metabolite profiles that correlate with microbial gene expression and host inflammatory markers. Longitudinal modeling indicates that effective therapy induces partial restoration of baseline microbiome composition and function, but residual dysbiosis persists in several conditions, suggesting a need for personalized microbiome-targeted interventions. Integrated host–microbiome networks highlight key hub taxa and host genes that may serve as therapeutic leverage points, such as modulators of innate immunity and barrier repair pathways. The study also uncovers potential predictive biomarkers for treatment response and relapse risk, enabling stratified management plans. By providing a comprehensive, multi-omics atlas of cutaneous microbial ecology across dermatological diseases, this work advances mechanistic understanding of microbe–host interactions in skin health and disease and lays groundwork for precision dermatology strategies that incorporate microbiome modulation as a therapeutic axis.

Project Overview

What This Project Is About

---

This project explores how the skin’s microbial communities change during skin diseases and how different data types can help us understand those changes. It combines observations of microbes (like bacteria and fungi) with other biological information to build a comprehensive picture of skin health and disease.

---

---

The Problem It Addresses

---

There is limited understanding of how microbial communities on the skin interact with the host during disease. Traditional studies look at one data type at a time, which can miss important interactions. This project aims to integrate multiple data sources to reveal patterns that could improve diagnosis and treatment.

---

---

Objectives of the Project

---

1. Describe how skin microbiome composition changes across common dermatological diseases.
2. Integrate multi-omics data (microbiome, host gene expression, and clinical information) to identify key interaction networks.
3. Identify microbial signatures associated with disease severity and treatment response.
4. Develop an approachable visualization of microbiome dynamics for clinicians.
5. Evaluate potential biomarkers for future diagnostic use.

---

---

What You Will Do Step by Step

---

1. Review background literature on skin microbiome and multi-omics.
2. Collect or access publicly available datasets with microbiome and host data from skin disease cohorts.
3. Preprocess data: quality control, normalization, and alignment of datasets.
4. Analyze microbiome diversity and composition across conditions.
5. Apply integrative analysis methods to combine datasets (e.g., network analysis, multi-omics fusion).
6. Identify microbial and host features linked to disease states.
7. Validate findings using cross-validation or external datasets.
8. Present results with simple visuals for non-experts and draft a short report.

---

---

Expected Outcome

Deliverables include a set of microbial signatures tied to diseases, an integrative model linking microbes to host responses, and user-friendly visuals for clinicians. The project could suggest biomarkers for future testing and highlight how combining data types improves understanding of skin diseases.