From theory to bench experiment by computer-assisted drug design

Area of Science:

• Drug discovery and development
• Computational chemistry
• Medicinal chemistry

Background:

• Optimizing drugs for multiple properties requires integrating computational design with practical medicinal chemistry.
• Interdisciplinary research combining pharmaceutical chemistry and computer science is crucial for advancing drug discovery.

Purpose of the Study:

• To present the activities and projects of an interdisciplinary research group at ETH Zürich focused on computer-assisted molecular design.
• To highlight the application of machine learning in hit and lead structure identification through virtual screening and de novo design.
• To introduce the concept of 'adaptive fitness landscapes' in drug discovery.

Main Methods:

• Utilizing machine learning algorithms for virtual screening and de novo drug design.
• Applying the 'adaptive fitness landscapes' concept to guide drug discovery projects.
• Integrating computational approaches with experimental medicinal chemistry validation.

Main Results:

• Demonstrated the potential of machine learning in accelerating hit and lead identification.
• Showcased practical examples of computer-assisted molecular design in drug discovery projects.
• Illustrated the utility of 'adaptive fitness landscapes' for optimizing multiple drug properties.

Conclusions:

• Tight integration of computational design and medicinal chemistry is essential for next-generation drug discovery.
• Machine learning and adaptive fitness landscapes offer powerful tools for efficient and optimized drug development.
• Interdisciplinary collaboration is vital for realizing the full potential of advanced molecular design strategies.