COReD Clinical Omics Resource for Respiratory-virus host factor Evidence and Discovery

sklrd Logo
Home
Evidence Matrix
Gene Search
Clinical scRNA-seq Host Genetics(GWAS) Cell-line Perturbation Animal Models
Cell Search
Clinical scRNA-seq Animal Models
Custom Analysis
Browse
Clinical scRNA-seq Host Genetics(GWAS) Perturbation Screens Animal Models
Downloads About Help

Help & Guide

🔬 Introduction

What is COReD?

COReD (Clinical Omics Resource for Respiratory-virus host factor Evidence and Discovery) is a clinical multi-omics resource for evaluating and discovering host factors in respiratory virus infections. It integrates patient single-cell transcriptomes, host genetics, functional perturbation screens and animal infection models to support gene-centered evidence review.

The central question in COReD is not only where a gene is expressed, but whether a candidate host factor has convergent, context-aware evidence across clinical samples, genetics, functional screens and infection models.

Who We Are

COReD is developed and maintained by the Zhao Lab at Guangzhou Medical University. Our research focuses on host-pathogen interactions and the identification of host factors that influence respiratory virus infection and disease progression.

For collaboration or data contribution inquiries, please contact: huqingtao@gzhmu.edu.cn

How to Use COReD

COReD supports four complementary routes for evaluating and discovering respiratory-virus host factor candidates. Choose the route that matches your starting point: a known gene, a matrix-wide screen, a cell type or a custom phenotype.

1. Candidate-first evidence review

If you already have a candidate host factor, start with Gene Search to inspect Host Genetics(GWAS), clinical scRNA-seq, animal-model and perturbation evidence.

Example genes: TMPRSS2, OAS1, MX1, IFITM3.

Start Search the official gene symbol from the home page or Gene Search.
Inspect Review evidence by virus, tissue, cell type and experimental layer.
Cross-check Use Evidence Matrix and Downloads to compare breadth and source datasets.

2. Matrix-guided prioritization

If you want to screen the database for broadly supported candidates, start from the Evidence Matrix.

Sort Rank genes by Tissue-Cell Context Breadth or Viral Spectrum Breadth.
Filter Focus on genes with selected evidence layers, such as Host Genetics(GWAS) plus Clinical scRNA.
Inspect Click matrix cells to jump into module-level evidence pages.

3. Cell-type-driven discovery

If you do not have a candidate gene, start from Cell Search, select a phenotype and key cell type, run DE, heatmap and enrichment, then return to Gene Search.

Choose context Select a virus, tissue and cell type, such as PBMC B cells or myeloid cells in an infection phenotype.
Discover Run DE, heatmap and enrichment to nominate candidate genes.
Validate context Return candidates to Gene Search and Evidence Matrix.

4. Custom cohort analysis

If default disease groups or studies do not match your question, use Custom Analysis to define sample groups and derive candidates from user-selected comparisons.

Select samples Filter by tissue, virus, study or phenotype and assign samples to group1/group2.
Analyze Run single-cell composition, DE, heatmap and enrichment workflows.
Prioritize Carry selected genes into Gene Search or Evidence Matrix.
Interpretation note: COReD prioritizes candidate genes and evidence contexts. Evidence breadth helps guide follow-up, but does not prove mechanism by itself.

📖 Tutorial

1. Gene Search (Home Page)

The home page provides a quick gene-centered entry point for host factor evidence review.

  1. Enter a gene name (e.g., OAS1, MX1, TNF). Case-insensitive.
  2. Click "Search" or press Enter.
  3. Use the resulting module pages to inspect clinical, functional, animal and genetics evidence where available.
  4. Use tissue, virus and cell-type views to understand context rather than treating a gene as globally supported.

2. About COReD and Evidence Matrix

The About page explains the database scope and design logic. The Evidence Matrix summarizes evidence availability and breadth for analyzable host-factor genes.

  1. Use About to understand the database scope and comparison with related resources.
  2. Use Evidence Matrix to identify which evidence layers and breadth categories can be inspected for a gene.
  3. Click matrix cells to continue into Host Genetics(GWAS), Clinical scRNA-seq, Animal Models or Cell-line Perturbation pages.

3. Clinical scRNA-seq Analysis

Navigate to Gene Search → Clinical scRNA-seq.

  1. Enter a gene name and click Search.
  2. Select a tissue type (PBMC, URT, BALF, Lung, etc.) from the top bar.
  3. Select a virus (SARS-CoV-2, Flu, etc.) from the left sidebar.
  4. Boxplots are generated for each cell type across severity groups:
    • HC (Healthy Control) — green
    • Mild — blue
    • Severe — red
  5. Statistical annotations: Mann-Whitney U test p-values (* p<0.05, ** p<0.01, *** p<0.001).

Download: Bulk download (PNG/PDF/CSV) via the "Download Chart" button, or use the ⋮ menu on any cell type card for individual export.

4. Host Genetics(GWAS) Analysis

Use Gene Search → Host Genetics(GWAS) to view regional host-genetics association results for the same gene.

5. Cell Line Data (KO / OE)

Navigate to Gene Search → Cell-line Perturbation.

  1. Enter a gene name and click Search.
  2. Switch between KO (Knockout) and OE (Overexpression) views.
  3. Heatmap View (default): Visualizes effects across all virus-cell line combinations.
  4. Scatter Plot View: Click a specific virus or cell line button for detailed data points.

6. Animal Model Data

Navigate to Gene Search → Animal Models. Similar to Clinical scRNA-seq, but organized by organism (Mouse, Hamster, Ferret) and tissue. The boxplot shows expression levels across experimental groups (e.g., infected vs. control).

7. Cell Search

Use Cell Search → Clinical scRNA-seq or Cell Search → Animal Models to inspect cell-type composition, expression context and downstream analysis for selected respiratory virus datasets. Cell Search can be followed by cell-type-specific differential expression, heatmap visualization and GO/KEGG enrichment analysis when available.

8. Custom Analysis

Navigate to the Custom Analysis page to define user-selected sample groups and run downstream single-cell analyses, including cell-type composition, differential expression, heatmap visualization and enrichment workflows.

9. Browse and Download

Use Browse for a visual overview of the four evidence layers: Clinical scRNA-seq, Host Genetics(GWAS), perturbation screens and animal models. Use Downloads to obtain the Evidence Matrix JSON/CSV exports, source-study metadata and reusable data tables for downstream analysis.

10. Quick Tips

❓ Frequently Asked Questions

What is COReD?
COReD is the Clinical Omics Resource for Respiratory-virus host factor Evidence and Discovery. It integrates patient single-cell transcriptomes, host genetics, functional screens and animal infection models for gene-centered host factor evidence review.
What data types are included?
What is the Host Factor Evidence Matrix?
The Evidence Matrix is a clickable evidence availability and breadth matrix for analyzable genes. Color intensity reflects evidence count or breadth, not a mechanistic proof score.
Why does COReD avoid a single host factor ranking score?
Different evidence types answer different questions. Clinical expression, Host Genetics(GWAS), functional screens and animal models should be interpreted by context. COReD therefore emphasizes traceable evidence categories instead of a subjective composite ranking.
How do I interpret the boxplots?
Each boxplot shows expression distribution per cell type. Box = IQR (Q1–Q3), line = median, whiskers = min/max. Significance is calculated via Mann-Whitney U test (* p<0.05, ** p<0.01, *** p<0.001). Y-axis shows normalized "Average UMI per 10k".
What does "Average UMI per 10k" mean?
UMIs (Unique Molecular Identifiers) per cell are normalized to 10,000 total UMIs to account for sequencing depth differences. The values shown are the cell-type average of this normalized metric within each sample.
KO vs. OE — what's the difference?
  • KO (Knockout): The gene is deleted to test if its loss affects viral infection.
  • OE (Overexpression): The gene is overexpressed to test if increased levels affect infection.
How can I cite COReD?
Cite as: "COReD: Clinical Omics Resource for Respiratory-virus host factor Evidence and Discovery" and include the database URL. Individual dataset users should cite the original publications listed on the Downloads page.
Is COReD freely accessible?
Yes. COReD is freely accessible for academic research. No registration required. Commercial use requires permission.
How will COReD be maintained?
COReD will be maintained under the current URL for at least five years. Source metadata, evidence indices and analysis modules will be updated as new respiratory-virus host-factor datasets become available.
How can I contribute data or report issues?
Contact huqingtao@gzhmu.edu.cn or use the Feedback page (💬 icon at the bottom-right of any page).

Still have questions?

We typically respond within 1–2 business days.

📦 Release & Version

2026-06-22 — v1.1.0 Latest
  • Released Evidence Matrix, Custom Analysis and expanded Download access.
  • Enabled streamlined DE, heatmap and enrichment workflows for clinical and animal cell-search modules.
2026-06-12 — v1.0.1
  • Improved Gene Search and Cell Search result pages with clearer boxplots and evidence-layer navigation.
  • Added admin feedback/statistics utilities and refined source metadata access.
2026-06-06 — v1.0.0
  • Initial launch of COReD as a respiratory-virus host-factor evidence resource.
  • Provided first gene-centered access to clinical scRNA-seq, Host Genetics(GWAS), perturbation and animal-model evidence.
Maintenance: COReD will be maintained under the current URL for at least five years, with evidence indices and source metadata updated as new respiratory-virus host-factor datasets become available.