Dimensions overall score 8.0

GitHub Code Pulse

No public code linked for this paper yet.

• DPD-Cancer achieves an Area Under ROC Curve (AUC) of up to 0.87 on strictly partitioned NCI60 data.result 90%
• DPD-Cancer achieves an AUC of up to 0.98 on ACLPred/MLASM datasets.result 90%
• For pGI50 prediction, DPD-Cancer achieved Pearson's correlation coefficients of up to 0.72 on independent test sets.result 90%
• DPD-Cancer utilizes a Graph Attention Transformer (GAT) framework for small molecule anti-cancer activity classification and quantitative prediction of cell-line specific responses.method 90%
• DPD-Cancer provides explainability by identifying and visualizing specific molecular substructures using its attention mechanism.technical 90%
• The model's performance may be limited in generalizability across other molecular sets due to reliance on the NCI60 dataset.limitation 80%
• DPD-Cancer uses chemistry-aware data partitioning to ensure evaluation on novel chemotypes.method 85%
• DPD-Cancer achieves an Area Under ROC Curve (AUC) of up to 0.87 on strictly partitioned NCI60 data.result 95%
• DPD-Cancer achieves an Area Under ROC Curve (AUC) of up to 0.98 on ACLPred/MLASM datasets.result 95%
• DPD-Cancer achieves Pearson's correlation coefficients of up to 0.72 for pGI50 prediction on independent test sets.result 95%
• DPD-Cancer utilizes a Graph Attention Transformer (GAT) framework for anti-cancer activity prediction.method 95%
• DPD-Cancer provides explainability by identifying and visualizing specific molecular substructures using its attention mechanism.technical 95%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card