This is the first in a three-part series discussing chemical diversity.
In Part 1, we introduce the concept of diversity in the context of fragment screening libraries.
How does one understand Chemical Diversity?
Merriam-Webster dictionary defines diversity as: “an instance of being composed of differing elements or qualities".
In the context of screening libraries, we refer to chemical diversity as the diversity of the chemical composition for a set of compounds. However, the approaches and criteria for chemical diversity can be as diverse as the libraries themselves!
Interestingly, despite the strong emphasis placed on library diversity, there is actually no correlation between chemical properties and biological activity reported in the literature (1).
Why is diversity a defining characteristic of chemical libraries?
From our point of view, diversity is directly related to the efficiency of the library. This includes coverage of chemical space. Chemical space is another topic that we will delve more into at a later date because it has wide interpretations, but for the purposes of this blog, it is understood as the theoretical space covered by all possible compounds with a defined set of properties. This ultimately relates to the diversity of hits obtained from the library and, in theory, a consistent hit-rate across diverse target classes.
What does Chemical Diversity Measure?
Common chemical diversity measures include chemical properties (e.g. molecular weight, clogP, hydrogen bonding), chemical fingerprints which are a binary representation of compound functionality and most recently, compound shape which includes so-named 3D fragments (2).
What will be covered next?
In Part 2 of our series, we will discuss the measures of diversity and the pros and cons of using them in fragment library design. We will conclude with Part 3, where we discuss our strategy for using diversity in design of our screening libraries.
To recap: Chemical diversity is a measure of the similarity of a collection of compounds although the defining diversity metric is not universal.
(1)- Martin et al., J Med Chem (2002) 45, 4350-4358