Candidate Biomarkers for NAFLD Identified via Bioinformatics and Machine Learning

Mingjie Guo's research dataset, published on figshare in 2026, identifies candidate genes associated with Non-alcoholic Fatty Liver Disease (NAFLD). The study employs a multi-omics strategy using datasets from the Gene Expression Omnibus (GEO) and machine learning models to pinpoint four hub biomarkers: MYO7A, AGTR1, IL1RN, and IGFBP2. The dataset is 9.4 MB in size and is shared under a CC-BY-4.0 license.

Use Cases

• Validating candidate biomarkers for NAFLD based on the identified hub genes MYO7A, AGTR1, IL1RN, and IGFBP2.
• Training machine learning models for NAFLD prediction using circulating insulin degradation-associated genes.
• Conducting systems biology analyses, such as weighted gene co-expression network analysis (WGCNA), on NAFLD-related gene expression data.
• Interpreting model predictions for NAFLD using SHAP (Shapley Additive Explanations) on the identified gene contributions.

Strengths

• Identifies four specific hub biomarkers (MYO7A, AGTR1, IL1RN, IGFBP2) for NAFLD.
• Employs a multi-omics strategy incorporating external datasets from the Gene Expression Omnibus (GEO).
• Uses multiple validation methods, including external datasets, cellular models, and animal models.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset is a 9.4 MB DOCX file, which may require parsing to extract structured data.

Provenance

Source

Gene Expression Omnibus (GEO) collection

Collection Method

Multi-omics strategy incorporating bioinformatics analysis and machine learning models.

Freshness

Last updated 2026-05-12 04:19:14; freshness should be verified.