MD-Syn
Synergistic drug combination prediction based on the multidimensional feature fusion method and attention mechanisms

MD-Syn is a user-friendly web portal for predicting drug synergy. It integrates a one-dimensional feature embedding module (1D-FEM), a two-dimensional feature embedding module (2D-FEM), and a deep neural network classifier to identify synergistic drug combinations. This interpretable framework prioritizes drug pairs by combining chemical structures and cancer cell line gene expression profiles, offering researchers actionable insights for combination therapy development.

How to Use This Tool

Step 1: Enter Drug Information

Enter the name of your drug in the 'Drug1 Name' field. This can be any standard drug name (e.g., Doxorubicin).

Step 2: Input SMILES Structure

Provide the SMILES notation of your drug in the 'Drug1 SMILES' field. SMILES is a chemical notation that represents the structure of your molecule.

Step 3: Select Cell Line

Choose the appropriate cell line from the dropdown menu. This selection will determine the context of your drug synergy prediction.

Step 4: Generate Prediction

Click the 'Predict Synergy' button to run the analysis. The tool will evaluate potential drug combinations and their synergistic effects.

Step 5: Receive Results

Enter your email address and click 'Send Results' to receive a detailed report of the predictions.

Interactive Plot Features:
  • Hover over bars to see exact probability values
  • Double-click on legend items to isolate specific data
  • Click and drag to zoom into specific regions
  • Use the toolbar in the top-right to:
    • Pan across the plot
    • Box select data points
    • Zoom in and out
    • Reset the view
    • Download the plot as a PNG file

Note: For best results, ensure that the SMILES notation is correct and properly formatted. You can find SMILES structures in chemical databases like PubChem or ChEMBL.

Drug Synergy Prediction Tool

Drug Synergy Probability Distribution

Prediction Results

© 2024 MD-Syn. All rights reserved.