Machine Learning-Determined Sex-Related Bladder Cancer Biomarkers from RNA-Seq Data

8.6 KB of gene panel data identifies sex-specific biomarkers for bladder cancer. The dataset results from applying four machine learning feature selection methods to gender-stratified RNA-seq data, achieving areas under the ROC curve of 0.932 and 0.914 for male and female panels, respectively. Authored by Joseph R. Pizzi and last updated on 2026-04-29, it is shared under a CC-BY-4.0 license.

Use Cases

• Training diagnostic classifiers for male bladder cancer based on the 50-gene panel associated with PI3K-AKT signaling.
• Training diagnostic classifiers for female bladder cancer based on the gene panel linked to extracellular matrix reorganization.
• Investigating the role of genes like PRAC1 and PCDH11Y in male tumor development as identified by the study.
• Exploring aberrant androgen signaling in female tumors using genes such as androgen receptor, PLXNA1, USP54, and PMEPA1 from the panel.

Strengths

• Performance metrics for the gene panels are provided, with AUC scores of 0.932 and 0.914.
• The dataset is derived from applying four distinct machine learning feature selection methods, which may increase robustness.
• The data is openly shared under a permissive CC-BY-4.0 license.

Limitations

• The dataset is very small at 8.6 KB, indicating limited scope.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

figshare

Collection Method

Machine learning analysis of gender and disease-stratified RNA-seq data using differential expression analysis, recursive feature elimination, logistic regression, and random forest.

Freshness

Last updated 2026-04-29 05:58:44; freshness should be verified.