DLAM: Deep Learning Predictions for Essential Proteins

Shunxian Zhou's research dataset, last updated April 2026, presents predictions for essential proteins using a deep learning model. The model integrates domain composition, subcellular localization, orthology, and gene expression with a protein-protein interaction network. Performance metrics, including ROC-AUC, AP, and F1-score, are reported from evaluations on the DIP and BioGRID datasets.

Use Cases

• Benchmarking new essential protein prediction models based on reported ROC-AUC and AP metrics.
• Training classifiers using multi-source biological features like domain composition and gene expression.
• Analyzing protein-protein interaction networks weighted by biological cues for topological insights.
• Studying model performance under class imbalance conditions as discussed in the description.

Strengths

• Model performance is quantitatively reported with metrics like ROC-AUC and AP.
• Evaluation was conducted on two established biological datasets, DIP and BioGRID.
• Data is openly licensed under CC-BY-4.0.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset is very small at 23.0 KB, indicating limited scope.

Provenance

Source

figshare, author Shunxian Zhou.

Collection Method

Likely contains model predictions and evaluation metrics generated by the DLAM deep learning framework.

Time Range

null

Freshness

Last updated 2026-04-14 22:01:28; freshness should be verified.

Geography

null