types of screening libraries in drug discovery

Traditionally, compound libraries consisted of compounds synthesized in previous research programs. Since these are drug research programs, the molecules tend to overlap with drug-like space – around 30 heavy atoms and a molecular weight that is often in the range of 400 to 500 g/mol range. Losartan, to the right, is a drug for high blood pressure and fits the profile of being drug-like. It has 30 heavy atoms (that’s the number of atoms excluding hydrogens) and a molecular weight of around 420. The level of potency for molecules found through screening a traditional library might be in the 1-to-10 micromolar range. In big pharma, the company collection will have over a million molecules, likely multiple milllions of compounds. Often, a subset of the full library will be used routinely in the screening. The company library is an important and valuable asset for finding actives at the start of a drug campaign.
Fragment libraries contain smaller molecules. Fragments are literally parts of a full drug with molecular weights often in the range of 200 to 300 g/mol. The size of a library is often around 3,000 compounds – so way smaller than a traditional, drug-like library. Because the molecules are smaller, the acceptable potency cut-off is higher at up to 1 millimolar. During optimization, leads from a fragment library need to be grown and expanded considerably to boost potency and other properties. The use of a fragment library falls within the umbrella of FBDD – fragment-based drug discovery. How does this work? To the right is a representative fragment library molecule with a molecular weight of about 200 g/mol and just 15 heavy atoms. To the left is our drug space. The fragment is not part of drug space, but very many molecules in drug space might contain this fragment as part of their structure. So, this one fragment probes the potential activity of all those very many molecules. In this manner, a relatively small number of fragments can effectively sample all of drug space. This is great news for smaller companies which almost certainly do not have access to a multi-million compound library for screening.
DNA-encoded libraries, also called DELs, involve linking a segment of DNA to a library molecule. So, the molecule is able to bind to the target in the assay, and the DNA “tag” is used to identify which molecule has bound the target. DELs can be absolutely huge, even in the billions of molecules. Because of the way that DELs are prepared and screened, the identification stage (often called “deconvolution”) can be challenging. DELs are a relatively new library technology, and the area is growing and being improved. As an emerging field, DNA-encoded libraries are more commonly used by large pharmaceutical companies, which have the resources to work in the developing field.

[TINTINTINTIN-uj]

Just wonder when everyone saying ic50 below 10uM is a good starting point, what concentration of the enzyme is typical being use for that ic50 value? As ic50 is also dependence on the protein conception (if it is an enzyme assay)? I have a ic50 20um with an enzyme assay with 100nM protein, if I could decrease the protein concentration I might get it into the boarder line?