AI for targeted polypharmacology: The next frontier in drug discovery.

In drug discovery, targeted polypharmacology, i.e., targeting multiple molecular targets with a single drug, is redefining therapeutic design to address complex diseases. Pre-selected pharmacological profiles, as exemplified in kinase drugs, promise enhanced efficacy and reduced toxicity. Historically, many of such drugs were discovered serendipitously, limiting predictability and efficacy, but currently artificial intelligence (AI) offers a transformative solution. Machine learning and deep learning techniques enable modeling protein structures, generating novel compounds, and decoding their polypharmacological effects, opening an avenue for more systematic and predictive multi-target drug design. This review explores the use of AI in identifying synergistic co-targets and delineating them from anti-targets that lead to adverse effects, and then discusses advances in AI-enabled docking, generative chemistry, and proteochemometric modeling of proteome-wide compound interactions, in the context of polypharmacology. We also provide insights into challenges ahead.

Investigating the conformational landscape of AlphaFold2-predicted protein kinase structures.

Summary: Protein kinases are a family of signaling proteins, crucial for maintaining cellular homeostasis. When dysregulated, kinases drive the pathogenesis of several diseases, and are thus one of the largest target categories for drug discovery. Kinase activity is tightly controlled by switching through several active and inactive conformations in their catalytic domain. Kinase inhibitors have been designed to engage kinases in specific conformational states, where each conformation presents a unique physico-chemical environment for therapeutic intervention. Thus, modeling kinases across conformations can enable the design of novel and optimally selective kinase drugs. Due to the recent success of AlphaFold2 in accurately predicting the 3D structure of proteins based on sequence, we investigated the conformational landscape of protein kinases as modeled by AlphaFold2. We observed that AlphaFold2 is able to model several kinase conformations across the kinome, however, certain conformations are only observed in specific kinase families. Furthermore, we show that the per residue predicted local distance difference test can capture information describing structural flexibility of kinases. Finally, we evaluated the docking performance of AlphaFold2 kinase structures for enriching known ligands. Taken together, we see an opportunity to leverage AlphaFold2 models for structure-based drug discovery against kinases across several pharmacologically relevant conformational states. Availability and implementation: All code used in the analysis is freely available at https://github.com/Harmonic-Discovery/AF2-kinase-conformational-landscape.

Chemogenomic library design strategies for precision oncology, applied to phenotypic profiling of glioblastoma patient cells.

Designing a targeted screening library of bioactive small molecules is a challenging task since most compounds modulate their effects through multiple protein targets with varying degrees of potency and selectivity. We implemented analytic procedures for designing anticancer compound libraries adjusted for library size, cellular activity, chemical diversity and availability, and target selectivity. The resulting compound collections cover a wide range of protein targets and biological pathways implicated in various cancers, making them widely applicable to precision oncology. We characterized the compound and target spaces of the virtual libraries, in comparison with a minimal screening library of 1,211 compounds for targeting 1,386 anticancer proteins. In a pilot screening study, we identified patient-specific vulnerabilities by imaging glioma stem cells from patients with glioblastoma (GBM), using a physical library of 789 compounds that cover 1,320 of the anticancer targets. The cell survival profiling revealed highly heterogeneous phenotypic responses across the patients and GBM subtypes.

Crowdsourced mapping of unexplored target space of kinase inhibitors.

Despite decades of intensive search for compounds that modulate the activity of particular protein targets, a large proportion of the human kinome remains as yet undrugged. Effective approaches are therefore required to map the massive space of unexplored compound-kinase interactions for novel and potent activities. Here, we carry out a crowdsourced benchmarking of predictive algorithms for kinase inhibitor potencies across multiple kinase families tested on unpublished bioactivity data. We find the top-performing predictions are based on various models, including kernel learning, gradient boosting and deep learning, and their ensemble leads to a predictive accuracy exceeding that of single-dose kinase activity assays. We design experiments based on the model predictions and identify unexpected activities even for under-studied kinases, thereby accelerating experimental mapping efforts. The open-source prediction algorithms together with the bioactivities between 95 compounds and 295 kinases provide a resource for benchmarking prediction algorithms and for extending the druggable kinome.

Chemogenomic Analysis of the Druggable Kinome and Its Application to Repositioning and Lead Identification Studies.

Owing to the intrinsic polypharmacological nature of most small-molecule kinase inhibitors, there is a need for computational models that enable systematic exploration of the chemogenomic landscape underlying druggable kinome toward more efficient kinome-profiling strategies. We implemented VirtualKinomeProfiler, an efficient computational platform that captures distinct representations of chemical similarity space of the druggable kinome for various drug discovery endeavors. By using the computational platform, we profiled approximately 37 million compound-kinase pairs and made predictions for 151,708 compounds in terms of their repositioning and lead molecule potential, against 248 kinases simultaneously. Experimental testing with biochemical assays validated 51 of the predicted interactions, identifying 19 small-molecule inhibitors of EGFR, HCK, FLT1, and MSK1 protein kinases. The prediction model led to a 1.5-fold increase in precision and 2.8-fold decrease in false-discovery rate, when compared with traditional single-dose biochemical screening, which demonstrates its potential to drastically expedite the kinome-specific drug discovery process.

Drug Target Commons 2.0: a community platform for systematic analysis of drug-target interaction profiles.

Drug Target Commons (DTC) is a web platform (database with user interface) for community-driven bioactivity data integration and standardization for comprehensive mapping, reuse and analysis of compound-target interaction profiles. End users can search, upload, edit, annotate and export expert-curated bioactivity data for further analysis, using an application programmable interface, database dump or tab-delimited text download options. To guide chemical biology and drug-repurposing applications, DTC version 2.0 includes updated clinical development information for the compounds and target gene-disease associations, as well as cancer-type indications for mutant protein targets, which are critical for precision oncology developments.

Drug Target Commons: A Community Effort to Build a Consensus Knowledge Base for Drug-Target Interactions.

Knowledge of the full target space of bioactive substances, approved and investigational drugs as well as chemical probes, provides important insights into therapeutic potential and possible adverse effects. The existing compound-target bioactivity data resources are often incomparable due to non-standardized and heterogeneous assay types and variability in endpoint measurements. To extract higher value from the existing and future compound target-profiling data, we implemented an open-data web platform, named Drug Target Commons (DTC), which features tools for crowd-sourced compound-target bioactivity data annotation, standardization, curation, and intra-resource integration. We demonstrate the unique value of DTC with several examples related to both drug discovery and drug repurposing applications and invite researchers to join this community effort to increase the reuse and extension of compound bioactivity data.

Improving the efficacy-safety balance of polypharmacology in multi-target drug discovery.

INTRODUCTION: Polypharmacology has emerged as an essential paradigm for modern drug discovery process. Multiple lines of evidence suggest that agents capable of modulating multiple targets in a selective manner may offer also improved balance between therapeutic efficacy and safety compared to single-targeted agents. Areas covered: Herein, the authors review the recent progress made in experimental and computational strategies for addressing the critical challenges with rational discovery of selective multi-targeted agents within the context of polypharmacological modelling. Specific focus is placed on multi-targeted mono-therapies, although examples of combinatorial polytherapies are also covered as an important part of the polypharmacology paradigm. The authors focus mainly on anti-cancer treatment applications, where polypharmacology is playing a key role in determining the efficacy-toxicity trade-off of multi-targeting strategies. Expert opinion: Even though it is widely appreciated that complex polypharmacological interactions can contribute both to therapeutic and adverse side-effects, systematic approaches for improving this balance by means of integrated experimental-computational strategies are still lacking. Future developments will be needed for comprehensive collection and harmonization of systems-wide target selectivity data, enabling better utilization and control for multi-targeted activities in the drug development process. Additional areas of future developments include model-based strategies for drug combination screening and improved pre-clinical validation options with animal models.

Computational-experimental approach to drug-target interaction mapping: A case study on kinase inhibitors.

Due to relatively high costs and labor required for experimental profiling of the full target space of chemical compounds, various machine learning models have been proposed as cost-effective means to advance this process in terms of predicting the most potent compound-target interactions for subsequent verification. However, most of the model predictions lack direct experimental validation in the laboratory, making their practical benefits for drug discovery or repurposing applications largely unknown. Here, we therefore introduce and carefully test a systematic computational-experimental framework for the prediction and pre-clinical verification of drug-target interactions using a well-established kernel-based regression algorithm as the prediction model. To evaluate its performance, we first predicted unmeasured binding affinities in a large-scale kinase inhibitor profiling study, and then experimentally tested 100 compound-kinase pairs. The relatively high correlation of 0.77 (p < 0.0001) between the predicted and measured bioactivities supports the potential of the model for filling the experimental gaps in existing compound-target interaction maps. Further, we subjected the model to a more challenging task of predicting target interactions for such a new candidate drug compound that lacks prior binding profile information. As a specific case study, we used tivozanib, an investigational VEGF receptor inhibitor with currently unknown off-target profile. Among 7 kinases with high predicted affinity, we experimentally validated 4 new off-targets of tivozanib, namely the Src-family kinases FRK and FYN A, the non-receptor tyrosine kinase ABL1, and the serine/threonine kinase SLK. Our sub-sequent experimental validation protocol effectively avoids any possible information leakage between the training and validation data, and therefore enables rigorous model validation for practical applications. These results demonstrate that the kernel-based modeling approach offers practical benefits for probing novel insights into the mode of action of investigational compounds, and for the identification of new target selectivities for drug repurposing applications.

C-SPADE: a web-tool for interactive analysis and visualization of drug screening experiments through compound-specific bioactivity dendrograms.

The advent of polypharmacology paradigm in drug discovery calls for novel chemoinformatic tools for analyzing compounds' multi-targeting activities. Such tools should provide an intuitive representation of the chemical space through capturing and visualizing underlying patterns of compound similarities linked to their polypharmacological effects. Most of the existing compound-centric chemoinformatics tools lack interactive options and user interfaces that are critical for the real-time needs of chemical biologists carrying out compound screening experiments. Toward that end, we introduce C-SPADE, an open-source exploratory web-tool for interactive analysis and visualization of drug profiling assays (biochemical, cell-based or cell-free) using compound-centric similarity clustering. C-SPADE allows the users to visually map the chemical diversity of a screening panel, explore investigational compounds in terms of their similarity to the screening panel, perform polypharmacological analyses and guide drug-target interaction predictions. C-SPADE requires only the raw drug profiling data as input, and it automatically retrieves the structural information and constructs the compound clusters in real-time, thereby reducing the time required for manual analysis in drug development or repurposing applications. The web-tool provides a customizable visual workspace that can either be downloaded as figure or Newick tree file or shared as a hyperlink with other users. C-SPADE is freely available at http://cspade.fimm.fi/.

Rationally designed transmembrane peptide mimics of the multidrug transporter protein Cdr1 act as antagonists to selectively block drug efflux and chemosensitize azole-resistant clinical isolates of Candida albicans.

Drug-resistant pathogenic fungi use several families of membrane-embedded transporters to efflux antifungal drugs from the cells. The efflux pump Cdr1 (Candida drug resistance 1) belongs to the ATP-binding cassette (ABC) superfamily of transporters. Cdr1 is one of the most predominant mechanisms of multidrug resistance in azole-resistant (AR) clinical isolates of Candida albicans. Blocking drug efflux represents an attractive approach to combat the multidrug resistance of this opportunistic human pathogen. In this study, we rationally designed and synthesized transmembrane peptide mimics (TMPMs) of Cdr1 protein (Cdr1p) that correspond to each of the 12 transmembrane helices (TMHs) of the two transmembrane domains of the protein to target the primary structure of the Cdr1p. Several FITC-tagged TMPMs specifically bound to Cdr1p and blocked the efflux of entrapped fluorescent dyes from the AR (Gu5) isolate. These TMPMs did not affect the efflux of entrapped fluorescent dye from cells expressing the Cdr1p homologue Cdr2p or from cells expressing a non-ABC transporter Mdr1p. Notably, the time correlation of single photon counting fluorescence measurements confirmed the specific interaction of FITC-tagged TMPMs with their respective TMH. By using mutant variants of Cdr1p, we show that these TMPM antagonists contain the structural information necessary to target their respective TMHs of Cdr1p and specific binding sites that mediate the interactions between the mimics and its respective helix. Additionally, TMPMs that were devoid of any demonstrable hemolytic, cytotoxic, and antifungal activities chemosensitize AR clinical isolates and demonstrate synergy with drugs that further improved the therapeutic potential of fluconazole in vivo.