Potency predictions are popular in compound design and optimization but are complicated by intrinsic limitations. Moreover, even for nonlinear methods, activity cliffs (ACs, formed by structural ...analogues with large potency differences) represent challenging test cases for compound potency predictions. We have devised a new test system for potency predictions, including AC compounds, that is based on partitioned matched molecular pairs (MMP) and makes it possible to monitor prediction accuracy at the level of analogue pairs with increasing potency differences. The results of systematic predictions using different machine learning and control methods on MMP-based data sets revealed increasing prediction errors when potency differences between corresponding training and test compounds increased, including large prediction errors for AC compounds. At the global level, these prediction errors were not apparent due to the statistical dominance of analogue pairs with small potency differences. Test compounds from such pairs were accurately predicted and determined the observed global prediction accuracy. Shapley value analysis, an explainable artificial intelligence approach, was applied to identify structural features determining potency predictions using different methods. The analysis revealed that numerical predictions of different regression models were determined by features that were shared by MMP partner compounds or absent in these compounds, with opposing effects. These findings provided another rationale for accurate predictions of similar potency values for structural analogues and failures in predicting the potency of AC compounds.
Display omitted
Given the increasing quest for selective kinase inhibitors, we have systematically investigated structural and structure-promiscuity relationships between promiscuous kinase ...inhibitors and other types with increasing potential for selective kinase inhibition. Therefore, inhibitors with different modes of action were extracted from X-ray structures of kinase complexes. For more than 18,000 promiscuous kinase inhibitors and 1253 type I1/2, II, and allosteric inhibitors with structurally confirmed mechanisms, analogue space was systematically charted. These inhibitors were active against a total of 426 human kinases. While nearly 80% of the promiscuous inhibitors formed related analogues series, only ~30% of other types of inhibitors were involved in such structural relationships and many of these inhibitors also had multi-kinase activity. Thus, most of the investigated type I1/2, II, and allosteric inhibitors with reported single-kinase activity were distinguished from promiscuous inhibitors, thus indicating potential for kinase selectivity. Structural relationships between promiscuous inhibitors and the subset of other inhibitors were organized in a matrix format including kinase activity profiles, revealing structure-promiscuity relationships for follow-up investigations.
Computational approaches that 'dock' small molecules into the structures of macromolecular targets and 'score' their potential complementarity to binding sites are widely used in hit identification ...and lead optimization. Indeed, there are now a number of drugs whose development was heavily influenced by or based on structure-based design and screening strategies, such as HIV protease inhibitors. Nevertheless, there remain significant challenges in the application of these approaches, in particular in relation to current scoring schemes. Here, we review key concepts and specific features of small-molecule-protein docking methods, highlight selected applications and discuss recent advances that aim to address the acknowledged limitations of established approaches.
Spirocycles frequently occur in natural products and experience increasing interest in drug discovery, given their richness in sp3 centers and distinct three‐dimensionality. We have systematically ...explored chemical space populated with currently available bioactive spirocycles. Compounds containing spiro systems were classified and their scaffolds and spirocyclic ring combinations analyzed. Nearly 47 000 compounds were identified that contained spirocycles in different structural contexts and were active against roughly 200 targets, among which several pharmaceutically relevant members of the G protein‐coupled receptor (GPCR) family were identified. Spirocycles and corresponding compounds displayed notable scaffold diversity but contained only limited numbers of combinations of differently sized rings. These observations indicate that there should be significant potential to further expand spirocyclic chemical space for drug discovery, exploiting the privileged substructure concept. Inspired by those findings, we embarked on the design and chemical synthesis of three distinct novel spirocyclic scaffolds that qualify for downstream library synthesis, thus exploring principally new chemical space with high potential for pharmaceutical research.
The spiral of life: The chemical space currently populated with available bioactive spirocycles has been systematically explored. Compounds containing spiro systems were classified and their scaffolds and spirocyclic ring combinations analyzed. Nearly 47 000 compounds were identified that contained spirocycles in different structural contexts and were active against roughly 200 targets. The design and chemical synthesis of three distinct novel spirocyclic scaffolds that qualify for downstream library synthesis was also performed.
Activity cliffs (ACs) are pairs of structurally similar or analogous active compounds with large differences in potency against the same target. For identifying and analyzing ACs, similarity and ...potency difference criteria must be determined and consistently applied. This can be done in various ways, leading to different types of ACs. In this work, we introduce a new category of ACs by combining different similarity criteria, including the formation of matched molecular pairs and structural isomer relationships. A systematic computational search identified such ACs in compounds with activity against a variety of targets. In addition to other ACs exclusively formed by structural isomers, the newly introduced category of ACs is rich in structure-activity relationship (SAR) information, straightforward to interpret from a chemical perspective, and further extends the current spectrum of ACs.
Similarity relationships. Shown are matched molecular pair (MMP) and structural isomer relationships, which provide the basis for the introduction of a new category of activity cliffs.
In qualitative or quantitative studies of structure–activity relationships (SARs), machine learning (ML) models are trained to recognize structural patterns that differentiate between active and ...inactive compounds. Understanding model decisions is challenging but of critical importance to guide compound design. Moreover, the interpretation of ML results provides an additional level of model validation based on expert knowledge. A number of complex ML approaches, especially deep learning (DL) architectures, have distinctive black-box character. Herein, a locally interpretable explanatory method termed Shapley additive explanations (SHAP) is introduced for rationalizing activity predictions of any ML algorithm, regardless of its complexity. Models resulting from random forest (RF), nonlinear support vector machine (SVM), and deep neural network (DNN) learning are interpreted, and structural patterns determining the predicted probability of activity are identified and mapped onto test compounds. The results indicate that SHAP has high potential for rationalizing predictions of complex ML models.