DLAM: Deep Learning Predictions for Essential Proteins

DLAM is a deep learning framework for predicting essential proteins, integrating domain composition, subcellular localization, orthology, and gene expression with a weighted protein-protein interaction network. The dataset contains model performance results, including ROC-AUC, AP, and F1-score, evaluated on the DIP and BioGRID datasets. It was authored by Shunxian Zhou and last updated on 2026-04 14.

Use Cases

• Benchmarking new essential protein prediction models based on reported ROC-AUC and AP metrics.
• Training attention-based deep learning models based on integrated biological cues like domain composition and subcellular localization.
• Analyzing the impact of multi-source biological information on prediction reliability under class imbalance.
• Comparing model performance against centrality measures and other deep learning methods like TCBB2021 and EPGAT.

Strengths

• Model performance is quantitatively reported with metrics like ROC-AUC and AP, including mean and standard deviation.
• Evaluation was conducted on two established biological datasets: DIP and the larger BioGRID dataset.
• Results suggest strong and stable discrimination and ranking capability across multiple folds.

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 98.1 KB, indicating a very limited scope, likely containing only summary results.

Provenance

Source

figshare, authored by Shunxian Zhou.

Collection Method

Results from a deep learning model (DLAM) for predicting essential proteins.

Time Range

null

Freshness

Last updated 2026-04-14 22:01:29; freshness should be verified.

Geography

null