LigTMap: ligand and structure-based target identification and activity prediction for small molecular compounds

Area of Science:

• Computational chemistry
• Drug discovery
• Bioinformatics

Background:

• Experimental drug target prediction is time-consuming and expensive.
• Efficient computational methods are needed to accelerate drug discovery.

Purpose of the Study:

• Introduce LigTMap, an automated online server for predicting protein targets of chemical compounds.
• Provide a valuable resource for computer-aided drug design and drug repurposing.

Main Methods:

• LigTMap integrates ligand similarity search with docking and binding similarity analysis.
• Utilizes 17 classes of therapeutic proteins from the PDBbind database.
• Workflow is fully automated for user convenience.

Main Results:

• Achieved over 70% success rate in predicting targets within the top-10 list for 1251 compounds.
• Demonstrated comparable performance to existing servers like SwissTargetPrediction and SEA.
• Showed improved top-10 success rates on newly compiled literature compounds.

Conclusions:

• LigTMap provides an efficient and accurate method for compound target prediction.
• The server offers ready-to-use ligand docking structures for further studies.
• LigTMap is freely accessible and its source code is available for reuse.