Design of novel antituberculosis compounds using graph-theoretical and substructural approaches

Area of Science:

• Medicinal Chemistry
• Computational Drug Design
• Infectious Diseases

Background:

• Rising resistance of Mycobacterium tuberculosis to current drugs poses a global health threat.
• Development of novel antituberculosis agents is crucial to combat drug-resistant strains.

Purpose of the Study:

• To design and predict novel antituberculosis agents using computational modeling.
• To evaluate the efficacy of two distinct modeling approaches for drug discovery.

Main Methods:

• Utilized a mixed approach combining fragment-based descriptors and topological substructural molecular design (TOPS-MODE).
• Employed a second model integrating two-dimensional (2D) and three-dimensional (3D) molecular descriptors.
• Analyzed a dataset of 167 compounds, including known antituberculosis agents and other pharmaceutical compounds.

Main Results:

• The TOPS-MODE based model achieved sensitivity, specificity, and accuracy above 80%.
• The 2D/3D descriptor model demonstrated comparable statistical indices exceeding 75%.
• Designed 12 imidazole structures with 100% accuracy using both models, validating the in silico methodology.

Conclusions:

• The developed in silico methodology is a promising tool for the rational design of effective antituberculosis drugs.
• The computational models can accelerate the discovery of new therapeutic agents against Mycobacterium tuberculosis.
• This approach aids in overcoming the challenge of drug resistance in tuberculosis treatment.