Related Experiment Videos
Focused combinatorial library design based on structural diversity, druglikeness and binding affinity score.
Gang Chen1, Suxin Zheng, Xiaomin Luo
1Drug Discovery and Design Center and State Key Laboratory of Drug Research, Shanghai Institute of Materia Medica, and Graduate School, Chinese Academy of Sciences, 555 Zuchongzhi Road, Shanghai 201203, PR China.
Journal of Combinatorial Chemistry
|May 10, 2005
Summary
This study introduces a new drug discovery approach using genetic algorithms (GAs) and improved scoring functions to create focused libraries. The method efficiently identifies novel drug candidates with high binding affinity and good ADME/T profiles.
Area of Science:
- Computational chemistry
- Drug discovery
- Medicinal chemistry
Background:
- Combinatorial chemistry challenges in drug discovery are mitigated by focused libraries and virtual screening.
- Genetic algorithms (GAs) are crucial for efficiently navigating vast chemical spaces in drug design.
- Scoring functions are pivotal for selecting promising molecules in targeted library generation.
Purpose of the Study:
- To develop a novel approach for generating target-focused drug libraries.
- To integrate structural diversity, binding affinity, and improved drug-likeness scoring.
- To introduce the LD1.0 software package implementing this new strategy.
Main Methods:
- Utilizing genetic algorithms (GAs) for efficient sampling of chemical space.
- Combining scores for structural diversity, binding affinity, and enhanced drug-likeness.
- Developing and applying the LD1.0 software for library generation.
- Validating the approach through cyclooxygenase (COX)2 and peroxisome proliferator-activated receptors gamma (PPARγ) focused library design.
Main Results:
- Successfully reproduced known COX2-selective inhibitors in a COX2-focused library.
- Generated novel, highly active PPARγ ligands, including key fragments of TZD drugs.
- Achieved results in approximately 15% of the time compared to traditional molecular docking.
- Demonstrated high hit rates, novel structures, and favorable ADME/T profiles.
Conclusions:
- The new approach offers an effective, reliable, and practical method for building focused drug libraries.
- LD1.0 software facilitates the creation of libraries with high hit rates and novel structures.
- This strategy significantly accelerates the drug discovery process while maintaining compound quality.