Subtyping of COVID-19 samples based on cell-cell interaction in single cell transcriptomes.
Kyeonghun Jeong1, Yooeun Kim2, Jaemin Jeon2
1Interdisciplinary Program in Bioengineering, Seoul National University, Seoul, 08826, Republic of Korea.
Scientific Reports
|November 10, 2023
Summary
This study reveals novel COVID-19 patient subtypes by analyzing cell-cell interactions. A "Severe-like moderate" subtype highlights immune differences impacting disease progression.
Area of Science:
- Immunology
- Computational Biology
- Infectious Diseases
Background:
- Single-cell transcriptome analysis has identified COVID-19 severity biomarkers.
- Existing studies often overlook immunological heterogeneity within similar clinical severity levels.
Purpose of the Study:
- To investigate patient heterogeneity in COVID-19 using cell-cell interaction scores for novel subtype discovery.
- To validate clustering methods and correlate subtypes with clinical severity.
Main Methods:
- Employed sample-level clustering based on cell-cell interaction scores.
- Validated clustering using external datasets and compared with gene expression-based methods.
- Characterized identified subtypes using known COVID-19 severity biomarkers.
Main Results:
- Cell-cell interaction score-based clustering showed superior reproducibility and purity.
- Clusters strongly correlated with the WHO ordinal severity score.
- Discovered a "Severe-like moderate" subtype with moderate clinical but severe molecular features.
Conclusions:
- Cell-cell interaction analysis is crucial for understanding COVID-19 patient heterogeneity.
- The "Severe-like moderate" subtype accurately predicts progression from moderate to severe disease.
- This subtype identification offers new insights into COVID-19 pathogenesis and severity prediction.
