Machine Learning Model Performance Comparison for Pancreatic Cancer miRNA Signature

Yanfei Zhu's dataset from 2026 contains performance metrics for an interpretable 20-microRNA diagnostic model for pancreatic cancer. It includes results from training on 216 samples and external validation across 585 RNA-seq and 30 serum-based samples. The data is stored in a 5.5 KB XLS file.

Use Cases

• Benchmarking diagnostic model performance based on reported AUC, sensitivity, and specificity metrics.
• Analyzing feature importance for miRNA biomarkers based on SHAP analysis results mentioned in the description.
• Studying cross-platform validation of biomarker signatures across RNA-seq and qRT-PCR datasets.

Strengths

• Model was externally validated on 585 independent RNA-seq samples and 30 serum-based samples.
• Performance metrics are provided, including a cross-validation AUC of 0.87 and external validation AUCs ranging from 0.78 to 0.83.
• Dataset is licensed under CC-BY-4.0, facilitating open reuse.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Data is based on retrospective datasets, as noted in the description.

Provenance

Source

figshare, author Yanfei Zhu

Collection Method

Model development and validation using publicly available datasets (GSE59856, GSE85589, TCGA-PAAD, GTEx, GSE128508).

Time Range

null

Freshness

Last updated 2026-05-04 17:37:47; freshness should be verified.

Geography

null