The essential roles of chemistry in high-throughput screening triage

Abstract

It is increasingly clear that academic high-throughput screening (HTS) and virtual HTS triage suffers from a lack of scientists trained in the art and science of early drug discovery chemistry. Many recent publications report the discovery of compounds by screening that are most likely artifacts or promiscuous bioactive compounds, and these results are not placed into the context of previous studies. For HTS to be most successful, it is our contention that there must exist an early partnership between biologists and medicinal chemists. Their combined skill sets are necessary to design robust assays and efficient workflows that will weed out assay artifacts, false positives, promiscuous bioactive compounds and intractable screening hits, efforts that ultimately give projects a better chance at identifying truly useful chemical matter. Expertise in medicinal chemistry, cheminformatics and purification sciences (analytical chemistry) can enhance the post-HTS triage process by quickly removing these problematic chemotypes from consideration, while simultaneously prioritizing the more promising chemical matter for follow-up testing. It is only when biologists and chemists collaborate effectively that HTS can manifest its full promise.

Figure 1. Primary impact points for chemistry in high-throughput screening and early lead discovery

The location of impact points covering the pre- and post-HTS stages highlights the importance of forming early partnerships between chemists and biologists in order to increase the chances of project success. HTS: High-throughput screening; vHTS: Virtual high-throughput screening.

Figure 2. Libraries, their uses and the development of a quality screening set

Placement of representative libraries in this space is necessarily subject to debate. For example, virtual libraries of real compounds can be used in vHTS, and the GDB contains commercially available compounds. HTS: High-throughput screening; vHTS: Virtual high-throughput screening.

Figure 3. Alternative approaches to chemical matter other than high-throughput screening

PubChem and ChEMBL are examples of publicly available chemical databases [–43].

Figure 4. Sample post-high-throughput screening triage screening tree

The process begins with data hand-off from the HTS, from which a series of cheminformatics operations can be performed to standardize, filter and annotate the data. As compounds are progressed, more detailed and time-consuming (‘higher-level’) analyses and resource-intensive experiments can be performed. We note that the order and details of each operation can vary depending on multiple factors (e.g., target, assay methods, library size and composition, available resources, expertise, project timeline and data volume). For instance, some groups may elect to test all actives by dose–response and counterscreens, and then perform more detailed chemocentric and cheminformatic analyses with the resulting data. DR: Dose–response; H2L: Hit-to-lead; HTS: High-throughput screening; QC: Quality control.

Figure 5. Example of a cheminformatics-assisted core analysis during a post-high-throughput screening triage

See also Box 4. The annotated high-throughput screening data are from a PKA screen performed in the ITDD. (A) Singletons; (B–F) representative compounds from three core classes. (B) is an obvious PAINS (substructure alert bolded). Compound (C) is fasudil, a known kinase inhibitor. Compound (D) has a structure reminiscent of known kinase inhibitors (note that the core containing compound [D] has several examples of active and inactive analogs, a promising early pattern that shows potential for early SAR exploration and argues against many of these compounds being statistical false positives). Core13, which contains compounds (E & F), has a profile we typically view with skepticism (note that while [E] is deemed active, all remaining compounds in this core are inactive, including the close analog [F]). This example also shows that effective use of modern cheminformatics software allows for large amounts of complex data to be analyzed efficiently.

Figure 6. The natural history of PAINS compounds reported in a recent manuscript

Searches performed in SciFinder. ≥90%: Compounds that are greater than 90% similar; Biol: Compounds that were reported in a ‘biological study’ from a journal article [,–172]; Exact: Exact substructure found; Sources: Commercial sources of exact compound.