Table 7_Exploration of candidate genes associated with rare SNVs in pulmonary stenosis usi

185 sporadic pulmonary stenosis patients and 100 healthy controls were analyzed via whole-exome sequencing to identify rare pathogenic variants. Three machine learning algorithms—LASSO, random forest, and XGBoost—were applied to prioritize 17 candidate genes associated with the condition. The dataset, shared by Yuting Liu under a CC-BY-4.0 license, provides a basis for investigating the genetic architecture of this congenital heart disease.

Use Cases

• Prioritizing candidate disease genes based on rare variant burden tests and machine learning feature selection.
• Comparing the performance of LASSO, random forest, and XGBoost algorithms for genomic feature selection.
• Validating candidate gene lists through protein-protein interaction network analysis.
• Assessing gene expression of prioritized candidates in human pulmonary artery endothelial cells via RT-qPCR.

Strengths

• Data is derived from whole-exome sequencing of 185 patients and 100 controls, providing a case-control study design.
• Analysis employed three distinct machine learning algorithms (LASSO, RF, XGBoost) and conventional statistical tests for cross-validation.
• The study identified and prioritized a specific list of 17 candidate genes for further investigation.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for certain analyses.
• The 1.3 MB file size suggests a small, summary-level dataset rather than raw sequencing data.

Provenance

Source

figshare

Collection Method

Whole-exome sequencing data analyzed with pathogenicity-filtering strategies, gene-level burden tests, and machine learning.

Freshness

Last updated 2026-06-04 22:01:42; freshness should be verified.