Drug Candidates for Brain Disorders Identified via Graph Neural Network

Jiafang Li's dataset lists 10 potential drug candidates for brain disorders, predicted using the Multi-omics Graph Transformer Network (MOGT). The data is stored in a 9.4 KB XLSX file and was last updated on 2026-05-29. The predictions were derived from integrating high-risk gene predictions for Parkinson's disease with the CMAP database.

Use Cases

• Prioritizing drug candidates for experimental validation based on predicted high-risk genes.
• Benchmarking new graph neural network models for disease gene prediction.
• Studying the link between multi-omics data and drug mechanisms for brain disorders.

Strengths

• Includes one experimentally validated drug candidate (UK-356618) for Parkinson's disease.
• Predictions are based on a published method (MOGT) that outperformed other approaches for five brain disorders.
• Dataset is small (9.4 KB) and easily accessible in XLSX format.

Limitations

• The dataset is very small (9.4 KB), indicating a limited scope of 10 candidate drugs.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

Jiafang Li

Collection Method

Predictions generated by the Multi-omics Graph Transformer Network (MOGT) integrated with the CMAP database.

Freshness

Last updated 2026-05-29 18:00:14; freshness should be verified.