Mapping Ligand-Shape Space for Protein-Ligand Systems: Distinguishing Key-in-Lock and Hand-in-Glove Proteins.

Many of the recently developed methods to study the shape of molecules permit one conformation of one molecule to be compared to another conformation of the same or a different molecule: a relative shape. Other methods provide an absolute description of the shape of a conformation that does not rely on comparisons or overlays. Any absolute description of shape can be used to generate a self-organizing map (shape map) that places all molecular shapes relative to one another; in the studies reported here, the shape fingerprint and ultrafast shape recognition methods are employed to create such maps. In the shape maps, molecules that are near one another have similar shapes, and the maps for the 102 targets in the DUD-E set have been generated. By examining the distribution of actives in comparison with their physical-property-matched decoys, we show that the proteins of key-in-lock type (relatively rigid receptor and ligand) can be distinguished from those that are more of a hand-in-glove type (more flexible receptor and ligand). These are linked to known differences in protein flexibility and binding-site size.

Turbocharging Matched Molecular Pair Analysis: Optimizing the Identification and Analysis of Pairs.

We have applied the two most commonly used methods for automatic matched pair identification, obtained the optimum settings, and discovered that the two methods are synergistic. A turbocharging approach to matched pair analysis is advocated in which a first round (a conservative categorical approach that uses an analogy with coin flips, heads corresponding to an increase in a measured property, tails to a decrease, and a biased coin to a structural change that reliably causes a change in that property) provides the settings for a second round (which uses the magnitude of the change in properties). Increased chemical specificity allows reliable knowledge to be extracted from smaller sets of pairs, and an assay-specific upper limit can be placed on the number of pairs required before adequate sampling of variability has been achieved.

DNA-dependent protein kinase (DNA-PK) inhibitors: structure-activity relationships for O-alkoxyphenylchromen-4-one probes of the ATP-binding domain.

Introduction of an O-alkoxyphenyl substituent at the 8-position of the 2-morpholino-4H-chromen-4-one pharmacophore enabled regions of the ATP-binding site of DNA-dependent protein kinase (DNA-PK) to be probed further. Structure-activity relationships have been elucidated for inhibition of DNA-PK and PI3K (p110α), with N-(2-(cyclopropylmethoxy)-4-(2-morpholino-4-oxo-4H-chromen-8-yl)phenyl)-2-morpholinoacetamide 11a being identified as a potent and selective DNA-PK inhibitor (IC(50)=8 nM).

WizePairZ: a novel algorithm to identify, encode, and exploit matched molecular pairs with unspecified cores in medicinal chemistry.

An algorithm to automatically identify and extract matched molecular pairs from a collection of compounds has been developed, allowing the learning associated with each molecular transformation to be readily exploited in drug discovery projects. Here, we present the application to an example data set of 11 histone deacetylase inhibitors. The matched pairs were identified, and corresponding differences in activity and lipophilicity were recorded. These property differences were associated with the chemical transformations encoded in the SMIRKS reaction notation. The transformations identified a subseries with the optimal balance of these two parameters. Enumeration of a virtual library of compounds using the extracted transformations identified two additional compounds initially excluded from the analysis with an accurate estimation of their biological activity. We describe how the WizePairZ system can be used to archive and apply medicinal chemistry knowledge from one drug discovery project to another as well as identify common bioisosteres.

Chemists: AI Is Here; Unite To Get the Benefits.

The latest developments in artificial intelligence (AI) have arrived into an existing state of creative tension between computational and medicinal chemists. At their most productive, medicinal and computational chemists have made significant progress in delivering new therapeutic agents into the clinic. However, the relationship between these communities has the prospect of being weakened by application of oversimplistic AI methods that, if they fail to deliver, will reinforce unproductive prejudices. We review what can be learned from our history of integrating QSAR and structure-based methods into drug discovery. Now with synthesis and testing available as contract services, the environment for computational innovation has changed and we consider the impact this may have on the relationships in our disciplines. We discuss the current state of interdisciplinary communication and suggest approaches to bring the subdisciplines together in order to improve computational medicinal chemistry and, most importantly, deliver better medicines to the clinic faster.

Can we accelerate medicinal chemistry by augmenting the chemist with Big Data and artificial intelligence?

AI comes to lead optimization: medicinal chemistry in all disease areas can be accelerated by exploiting our pre-competitive knowledge in an unbiased way.

An Orally Bioavailable, Indole-3-glyoxylamide Based Series of Tubulin Polymerization Inhibitors Showing Tumor Growth Inhibition in a Mouse Xenograft Model of Head and Neck Cancer.

A number of indole-3-glyoxylamides have previously been reported as tubulin polymerization inhibitors, although none has yet been successfully developed clinically. We report here a new series of related compounds, modified according to a strategy of reducing aromatic ring count and introducing a greater degree of saturation, which retain potent tubulin polymerization activity but with a distinct SAR from previously documented libraries. A subset of active compounds from the reported series is shown to interact with tubulin at the colchicine binding site, disrupt the cellular microtubule network, and exert a cytotoxic effect against multiple cancer cell lines. Two compounds demonstrated significant tumor growth inhibition in a mouse xenograft model of head and neck cancer, a type of the disease which often proves resistant to chemotherapy, supporting further development of the current series as potential new therapeutics.

Matched molecular pair analysis in drug discovery.

Multiple parameter optimisation in drug discovery is difficult, but Matched Molecular Pair Analysis (MMPA) can help. Computer algorithms can process data in an unbiased way to yield design rules and suggest better molecules, cutting the number of design cycles. The approach often makes more suggestions than can be processed manually and methods to deal with this are proposed. However, there is a paucity of contextually specific design rules, which would truly make the technique powerful. By combining extracted information from multiple sources there is an opportunity to solve this problem and advance medicinal chemistry in a matter of months rather than years.