COMBAT: A Preprocessed Blood Atlas of COVID-19 Disease Severity

A preprocessed single-cell multi-omics blood dataset from the COMBAT Consortium, originally published in 2022. The 1.3 GB dataset has been processed with standard scanpy pipelines and includes dimensionality reductions like PCA, scVI, scANVI, and scPoli. It was uploaded to figshare by Eljas Roellin in May 2026 for convenient access.

Use Cases

• Identify cellular hallmarks of COVID-19 severity based on multi-omics blood profiling.
• Benchmark dimensionality reduction and integration methods like scVI and scANVI on a preprocessed dataset.
• Analyze immune cell population shifts associated with disease specificity mentioned in the original study.

Strengths

• Dataset is 1.3 GB in size, indicating substantial processed data.
• Includes multiple advanced dimensionality reductions (PCA, scVI, scANVI, scPoli) applied to the raw data.
• Derived from a peer-reviewed publication in Cell and a consortium data release on Zenodo.

Limitations

• Row count and specific column names are unknown, limiting suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.

Provenance

Source

COvid-19 Multi-omics Blood ATlas (COMBAT) Consortium via Zenodo and Kaggle.

Collection Method

Blood samples processed with single-cell multi-omics techniques, then preprocessed with scanpy.

Time Range

Data collection period is not specified in the provided metadata.

Freshness

Last updated 2026-05-04 16:11:24; freshness should be verified.

Geography

Geographic coverage is not specified in the provided metadata.