Gene Expression Data for Discriminating MASL and MASH Liver Disease

A 2026 study by Guo Wu Lin integrates transcriptomic data from 149 MASL and 158 MASH samples for training, and 51 MASL and 155 MASH samples for validation. The dataset contains results from an artificial neural network model identifying a six-gene signature for early disease discrimination, validated with clinical samples.

Use Cases

• Training diagnostic classifiers based on the identified six-gene signature (MMP9, FABP5, TREM2, CTSD, UBD, MAP2K1).
• Analyzing immune cell infiltration patterns in liver disease based on CIBERSORT results mentioned in the description.
• Validating gene expression markers for MASLD progression using the provided training and validation cohorts.
• Investigating cell-type-specific expression patterns using the single-cell RNA sequencing data referenced in the study.

Strengths

• Includes a validation cohort of 206 samples (51 MASL, 155 MASH) for model evaluation.
• The artificial neural network model achieved an AUC of 0.893 in the validation cohort.
• Results were validated with quantitative PCR on 60 clinical liver tissue samples.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Data may reflect bias inherent to the specific Gene Expression Omnibus datasets used.

Provenance

Source

Integrated from five datasets in the Gene Expression Omnibus (GEO).

Collection Method

Differential expression analysis, network construction, and machine learning feature selection applied to transcriptomic data.

Time Range

null

Freshness

Last updated 2026-04-20 04:11:45; freshness should be verified.

Geography

null