GABRA1 Missense Variants: Pathogenicity and Phenotype Analysis for 61 Variants

A dataset of 61 GABRA1 missense variants curated to evaluate computational tools for pathogenicity classification. It includes 30 pathogenic/likely pathogenic variants from patients and 31 benign/likely benign controls from the gnomAD database. The dataset was created by Wen-Hui Liu and last updated on 2026-05-18.

Use Cases

• Benchmarking pathogenicity prediction algorithms based on the evaluation of 34 computational tools described in the study
• Analyzing molecular subregional effects on phenotype based on variant domain location (e.g., transmembrane vs. extracellular)
• Training ensemble or deep learning models for variant classification using the curated set of labeled pathogenic and benign variants

Strengths

• Includes 61 clearly labeled variants (30 pathogenic/likely pathogenic, 31 benign/likely benign) enabling supervised learning tasks
• Performance metrics for 34 distinct algorithms are provided, including accuracy, AUC, and statistical significance (P-values)
• Variant analysis shows significant domain-dependent phenotype distribution (P = 0.01), adding biological context

Limitations

• Row count is unknown, which may limit suitability assessment
• Column-level documentation is absent; field semantics must be inferred after download
• The dataset is very small (7.6 KB), indicating limited scope and potential for model overfitting

Provenance

Source

figshare

Collection Method

Curated from published literature and the gnomAD database, as described in the study methods.

Freshness

Last updated 2026-05-18 22:02:41; freshness should be verified