Fragment-based screening has become increasingly popular over the last 15 years and has proven to be a viable alternative to high-throughput screening.

A little while ago I posed the question "Do you include chiral fragments in your screen?" And the results were interesting. Whilst many, many folks clicked on the link very few actually responded to the poll, perhaps something people are thinking about but don't have an answer?

For those that did respond it seems that very few actively exclude chiral fragments.

Of those that included chiral fragments the responses were pretty evenly split between those that include racemates and only worry about the enantiomers if they appear as a hit, those that only include racemates if the individual enantiomers are available and those that only include individual enantiomers.

One of the arguments against including chiral fragments has been it increases the complexity of the fragments and as the ligand/receptor match becomes more complex the probability of any given molecule matching falls. However, biological targets are almost invariably chiral and selectivity often relies on stereochemical features.

One of the key attractions of fragment screening is the ability to explore/validate hits without committing significant chemistry resources, so I’d argue that the ready availability of related analogues ought to be part of the library selection criterion and by extension having access to the individual enantiomers should be an important consideration.