Marcel L. Verdonk

General and targeted statistical potentials for protein–ligand interactions

We present a novel atom–atom potential derived from a database of protein–ligand complexes. First, we clarify the similarities and differences between two statistical potentials described in the literature, PMF and Drugscore. We highlight shortcomings caused by an important factor unaccounted for in their reference states, and describe a new potential, which we name the Astex Statistical Potential (ASP). ASPs reference state considers the difference in exposure of protein atom types towards ligand binding sites. We show that this new potential predicts binding affinities with an accuracy similar to that of Goldscore and Chemscore. We investigate the influence of the choice of reference state by constructing two additional statistical potentials that differ from ASP only in this respect. The reference states in these two potentials are defined along the lines of Drugscore and PMF. In docking experiments, the potential using the new reference state proposed for ASP gives better success rates than when these literature reference states were used; a success rate similar to the established scoring functions Goldscore and Chemscore is achieved with ASP. This is the case both for a large, general validation set of protein–ligand structures and for small test sets of actives against four pharmaceutically relevant targets. Virtual screening experiments for these targets show less discrimination between the different reference states in terms of enrichment. In addition, we describe how statistical potentials can be used in the construction of targeted scoring functions. Examples are given for cdk2, using four different targeted scoring functions, biased towards increasingly large target‐specific databases. Using these targeted scoring functions, docking success rates as well as enrichments are significantly better than for the general ASP scoring function. Results improve with the number of structures used in the construction of the target scoring functions, thus illustrating that these targeted ASP potentials can be continuously improved as new structural data become available. Proteins 2005.

Experiences in fragment-based drug discovery.

Fragment-based drug discovery (FBDD) has become established in both industry and academia as an alternative approach to high-throughput screening for the generation of chemical leads for drug targets. In FBDD, specialised detection methods are used to identify small chemical compounds (fragments) that bind to the drug target, and structural biology is usually employed to establish their binding mode and to facilitate their optimisation. In this article, we present three recent and successful case histories in FBDD. We then re-examine the key concepts and challenges of FBDD with particular emphasis on recent literature and our own experience from a substantial number of FBDD applications. Our opinion is that careful application of FBDD is living up to its promise of delivering high quality leads with good physical properties and that in future many drug molecules will be derived from fragment-based approaches.

Automated Protein–Ligand Crystallography for Structure‐Based Drug Design

An approach to automate protein–ligand crystallography is presented, with the aim of increasing the number of structures available to structure‐based drug design. The methods we propose deal with the automatic interpretation of diffraction data for targets with known protein structures, and provide easy access to the results. Central to the system is a novel procedure that fully automates the placement of ligands into electron density maps. Automation provides an objective way to structure solution, whereas manual placement can be rather subjective, especially for data of low to medium resolution. Ligands are placed by docking into electron density, whilst taking care of protein–ligand interactions. The ligand fitting procedure has been validated on both public domain and in‐house examples. Some of the latter deal with cocktails of low‐molecular weight compounds, as used in fragment‐based drug discovery by crystallography. For such library‐screening experiments we show that the method can automatically identify which of the compounds from a cocktail is bound.

Group efficiency: a guideline for hits-to-leads chemistry.

Herein we describe the concept of group efficiency (GE), which is an extension of ligand efficiency (LE), that we find particularly useful during the hit-tolead and optimisation stages of drug discovery projects. LE has already become popular amongst medicinal chemists. It is used to normalise the binding affinity of a compound with respect to its molecular weight and is a simple way to rank and compare the affinities of compounds with different sizes. The term LE was first suggested by Hopkins et al. as a measure of the free energy of binding (DGb) divided by the molecular size and is related to a publication from Kuntz et al. :

Detection of secondary binding sites in proteins using fragment screening

Significance The regulation of proteins in biological systems is essential to their function and nature has evolved a diverse array of mechanisms by which to achieve such regulation. Indeed, the primary function of a protein may be regulated by interaction with endogenous ligands or other protein partners binding at secondary sites. In this study, we report that fragment screening using X-ray crystallography can identify such secondary sites that may have a biological function, which in turn implies that the opportunities for modulating protein function with small molecules via such sites are far more widespread than previously assumed. Many of the secondary sites we discovered were previously unknown and therefore offer potential for novel approaches to modulate these protein targets. Proteins need to be tightly regulated as they control biological processes in most normal cellular functions. The precise mechanisms of regulation are rarely completely understood but can involve binding of endogenous ligands and/or partner proteins at specific locations on a protein that can modulate function. Often, these additional secondary binding sites appear separate to the primary binding site, which, for example for an enzyme, may bind a substrate. In previous work, we have uncovered several examples in which secondary binding sites were discovered on proteins using fragment screening approaches. In each case, we were able to establish that the newly identified secondary binding site was biologically relevant as it was able to modulate function by the binding of a small molecule. In this study, we investigate how often secondary binding sites are located on proteins by analyzing 24 protein targets for which we have performed a fragment screen using X-ray crystallography. Our analysis shows that, surprisingly, the majority of proteins contain secondary binding sites based on their ability to bind fragments. Furthermore, sequence analysis of these previously unknown sites indicate high conservation, which suggests that they may have a biological function, perhaps via an allosteric mechanism. Comparing the physicochemical properties of the secondary sites with known primary ligand binding sites also shows broad similarities indicating that many of the secondary sites may be druggable in nature with small molecules that could provide new opportunities to modulate potential therapeutic targets.

COSMIC-3D provides structural perspectives on cancer genetics for drug discovery

Nature GeNetics | VOL 50 | SEPTEMBER 2018 | 1198–1204 | www.nature.com/naturegenetics expression in each clone and cell), we find it misleading to describe clonal aRME as the “autosomal analog of X chromosome inactivation”1. Finally, Gimelbrant and colleagues suggest that single-cell RNA-seq may not be suitable (or may be underpowered) for allelic gene expression analyses. We would like to refer them to data from our study2, in which the significance of clonal aRME is reported per gene and chromosome. When XCI is used as an internal control in the cells, it is apparent that single-cell RNA-seq has sufficient power to identify essentially all genes on the X chromosome as having clonal monoallelic gene expression. Moreover, the single-cell RNA-seq in our study identified the same imprinted genes as bulk RNAseq performed on the same primary cells2, thus again confirming power and accuracy. The reference13 cited by Gimelbrant and colleagues regarding “challenges of singlecell RNA-seq, which make confident detection of less extreme bias difficult” simply reports that technical controls are needed to correctly adjust for the incomplete sensitivity of the libraries from each cell, procedures that we have advocated for and used in our studies2,9. Altogether, we agree with Gimelbrant and colleagues’ notion that clonal aRME occurs in mammalian cells beyond the classical cases such as antigen14 and olfactory15 receptors. However, it is now evident2 to us that this phenomenon is biologically much rarer than initially thought4–8,16, and it affects mostly genes that are weakly expressed, thus calling its functional role into question. Our study2 provided the first genome-wide insights into primary and in vivo cells, whereas previous bulk analyses on clonal aRME used immortalized cell lines4,5, which are notorious for genomic changes. (A full understanding of the clones used by Gimelbrant and colleagues in ref. 4, gained through modern methods such as RNA and DNA sequencing, has not been possible because of the authors’ unwillingness to share the cell clones.) In the future, we look forward to seeing more extensive in vivo datasets on clonal aRME, given that singlecell RNA-seq has become routine and can now be applied to resolve additional questions on monoallelic gene expression. ❐