Defining Levels of Automated Chemical Design

Area of Science:

• Computational chemistry
• Medicinal chemistry
• Drug discovery

Background:

• Computational methods are increasingly used for automated small molecule design in drug discovery.
• Existing literature lacks standardized terminology and key attributes to differentiate these automated design systems.
• This ambiguity hinders clear understanding and evaluation of various computational approaches.

Purpose of the Study:

• To introduce a novel framework, Automated Chemical Design (ACD) Levels, for classifying automated small molecule design systems.
• To define system autonomy based on ideation and decision-making capabilities.
• To provide a common language for the field of computational drug discovery.

Main Methods:

• Development of the Automated Chemical Design (ACD) Levels framework.
• Analysis of existing literature to identify and categorize automated small molecule design systems.
• Illustration of the framework using literature exemplars and notable methods.

Main Results:

• The proposed ACD Levels framework clearly defines the autonomy of automated chemical design systems.
• The framework categorizes various notable methods and applications within its defined levels.
• A common language is established for describing automated small molecule design systems.

Conclusions:

• The ACD Levels provide a standardized approach to classifying automated small molecule design systems.
• This framework enhances clarity and facilitates better understanding and evaluation for medicinal chemists.
• Adoption of ACD Levels will advance the field of computational drug discovery through consistent terminology.