Target fishing for chemical compounds using target-ligand activity data and ranking based methods

Area of Science:

• Computational Chemistry
• Drug Discovery
• Bioinformatics

Background:

• Identifying chemical compound targets is crucial for early drug discovery.
• Understanding selectivity, off-target effects, and toxicity relies on accurate target identification.
• Target fishing, identifying likely targets for a compound, is an active research area.

Purpose of the Study:

• To develop and present novel computational techniques for ranking potential targets of chemical compounds.
• To improve the prioritization of targets in drug discovery to identify compounds with higher activity.
• To enhance the understanding of compound-target interactions for better drug development.

Main Methods:

• Utilized extensions of Support Vector Machine (SVM) algorithms for target ranking.
• Employed ranking perceptron algorithms tailored for the target fishing problem.
• Conducted extensive experimental studies to validate the developed methods.

Main Results:

• The developed techniques effectively rank potential drug targets for chemical compounds.
• Methods show significant performance improvements over existing approaches.
• Outperformed previous methods by 2% to 60% across various evaluation criteria.

Conclusions:

• The novel computational methods enhance the accuracy and efficiency of target fishing.
• Improved target prioritization aids in identifying selective and less toxic drug candidates.
• These advancements contribute to more effective early-stage drug discovery processes.