CFD+CD14+ Monocytes in Myasthenia Gravis: Multi-Omics and Machine Learning Analysis

A multi-omics integration study identifies a potential pathogenic monocyte subset in myasthenia gravis. The dataset includes single-cell RNA sequencing data from peripheral blood of MG patients and healthy controls, integrated with GWAS summary data and expression quantitative trait loci. Shuang Li authored this work, which was last updated on 2026-05-01.

Use Cases

• Identify disease-associated cell subsets based on TWAS activity status across cell types.
• Train machine learning models for feature selection using the six signature genes (FKBP15, EHMT1, CHPT1, KLC1, SCPEP1, CFD).
• Analyze intercellular communication networks comparing CFD+ and CFD- CD14+ monocytes.
• Investigate pathway enrichment in monocyte subsets related to mTORC1 signaling, complement, and inflammatory response.

Strengths

• Integrates data from multiple public sources including GEO, GWAS catalog, and GTEx.
• Applies seven machine learning algorithms and uses a benchmark to select the best model.
• Includes a validation dataset generated from scRNA-seq on peripheral blood from MG patients and controls.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The primary data file is a 2.0 MB DOCX document, which may not be a standard data format for analysis.

Provenance

Source

figshare

Collection Method

Integrated multi-omics data from GEO, GWAS catalog, and GTEx databases; generated validation scRNA-seq data.

Freshness

Last updated 2026-05-01 05:25:22; freshness should be verified.