Profile-QSAR: a novel meta-QSAR method that combines activities across the kinase family to accurately predict affinity, selectivity, and cellular activity

The Development and Application of KinomePro-DL: A Deep Learning Based Online Small Molecule Kinome Selectivity Profiling Prediction Platform

Characterizing the kinome selectivity profiles of kinase inhibitors is essential in the early stages of novel small-molecule drug discovery. This characterization is critical for interpreting potential adverse events caused by off-target polypharmacology effects and provides unique pharmacological insights for drug repurposing development of existing kinase inhibitor drugs. However, experimental profiling of whole kinome selectivity is still time-consuming and resource-demanding. Here, we report a deep learning classification model using an in-house built data set of inhibitors against 191 well-representative kinases constructed based on a novel strategy by systematically cleaning and integrating six public data sets. This model, a multitask deep neural network, predicts the kinome selectivity profiles of compounds with novel structures. The model demonstrates excellent predictive performance, with auROC, prc-AUC, Accuracy, and Binary_cross_entropy of 0.95, 0.92, 0.90, and 0.37, respectively. It also performs well in a priori testing for inhibitors targeting different categories of proteins from internal compound collections, significantly improving over similar models on data sets from practical application scenarios. Integrated to subsequent machine learning-enhanced virtual screening workflow, novel CDK2 kinase inhibitors with potent kinase inhibitory activity and excellent kinome selectivity profiles are successfully identified. Additionally, we developed a free online web server, KinomePro-DL, to predict the kinome selectivity profiles and kinome-wide polypharmacology effects of small molecules (available on kinomepro-dl.pharmablock.com). Uniquely, our model allows users to quickly fine-tune it with their own training data sets, enhancing both prediction accuracy and robustness.

Data-oriented protein kinase drug discovery

The continued growth of data from biological screening and medicinal chemistry provides opportunities for data-driven experimental design and decision making in early-phase drug discovery. Approaches adopted from data science help to integrate internal and public domain data and extract knowledge from historical in-house data. Protein kinase (PK) drug discovery is an exemplary area where large amounts of data are accumulating, providing a valuable knowledge base for discovery projects. Herein, the evolution of PK drug discovery and development of small molecular PK inhibitors (PKIs) is reviewed, highlighting milestone developments in the field and discussing exemplary studies providing a basis for increasing data orientation of PK discovery efforts.

HyperPCM: Robust Task-Conditioned Modeling of Drug-Target Interactions

A central problem in drug discovery is to identify the interactions between drug-like compounds and protein targets. Over the past few decades, various quantitative structure-activity relationship (QSAR) and proteo-chemometric (PCM) approaches have been developed to model and predict these interactions. While QSAR approaches solely utilize representations of the drug compound, PCM methods incorporate both representations of the protein target and the drug compound, enabling them to achieve above-chance predictive accuracy on previously unseen protein targets. Both QSAR and PCM approaches have recently been improved by machine learning and deep neural networks, that allow the development of drug-target interaction prediction models from measurement data. However, deep neural networks typically require large amounts of training data and cannot robustly adapt to new tasks, such as predicting interaction for unseen protein targets at inference time. In this work, we propose to use HyperNetworks to efficiently transfer information between tasks during inference and thus to accurately predict drug-target interactions on unseen protein targets. Our HyperPCM method reaches state-of-the-art performance compared to previous methods on multiple well-known benchmarks, including Davis, DUD-E, and a ChEMBL derived data set, and particularly excels at zero-shot inference involving unseen protein targets. Our method, as well as reproducible data preparation, is available at https://github.com/ml-jku/hyper-dti.

Large-scale comparison of machine learning methods for profiling prediction of kinase inhibitors

Conventional machine learning (ML) and deep learning (DL) play a key role in the selectivity prediction of kinase inhibitors. A number of models based on available datasets can be used to predict the kinase profile of compounds, but there is still controversy about the advantages and disadvantages of ML and DL for such tasks. In this study, we constructed a comprehensive benchmark dataset of kinase inhibitors, involving in 141,086 unique compounds and 216,823 well-defined bioassay data points for 354 kinases. We then systematically compared the performance of 12 ML and DL methods on the kinase profiling prediction task. Extensive experimental results reveal that (1) Descriptor-based ML models generally slightly outperform fingerprint-based ML models in terms of predictive performance. RF as an ensemble learning approach displays the overall best predictive performance. (2) Single-task graph-based DL models are generally inferior to conventional descriptor- and fingerprint-based ML models, however, the corresponding multi-task models generally improves the average accuracy of kinase profile prediction. For example, the multi-task FP-GNN model outperforms the conventional descriptor- and fingerprint-based ML models with an average AUC of 0.807. (3) Fusion models based on voting and stacking methods can further improve the performance of the kinase profiling prediction task, specifically, RF::AtomPairs + FP2 + RDKitDes fusion model performs best with the highest average AUC value of 0.825 on the test sets. These findings provide useful information for guiding choices of the ML and DL methods for the kinase profiling prediction tasks. Finally, an online platform called KIPP ( https://kipp.idruglab.cn ) and python software are developed based on the best models to support the kinase profiling prediction, as well as various kinase inhibitor identification tasks including virtual screening, compound repositioning and target fishing.

Recent advances of artificial intelligence in melanoma clinical practice

Skin melanoma is a lethal cancer. The incidence of melanoma is increasing rapidly in all regions of the world. Despite significant breakthroughs in melanoma treatment in recent years, precise diagnosis of melanoma is still a challenge in some cases. Even specialized physicians may need time and effort to make accurate judgments. As artificial intelligence (AI) technology advances into medical practice, it may bring new solutions to this problem based on its efficiency, accuracy, and speed. This paper summarizes the recent progress of AI in melanoma-related applications, including melanoma diagnosis and classification, the discovery of new medication, guiding treatment, and prognostic assessment. The paper also compares the effectiveness of various algorithms in melanoma application and suggests future research directions for AI in melanoma clinical practice.

Artificial intelligence methods in kinase target profiling: Advances and challenges

Kinases have a crucial role in regulating almost the full range of cellular processes, making them essential targets for therapeutic interventions against various diseases. Accurate kinase-profiling prediction is vital for addressing the selectivity/specificity challenges in kinase drug discovery, which is closely related to lead optimization, drug repurposing, and the understanding of potential drug side effects. In this review, we provide an overview of the latest advancements in machine learning (ML)-based and deep learning (DL)-based quantitative structure-activity relationship (QSAR) models for kinase profiling. We highlight current trends in this rapidly evolving field and discuss the existing challenges and future directions regarding experimental data set construction and model architecture design. Our aim is to offer practical insights and guidance for the development and utilization of these approaches.

Drug discovery for parasitic diseases: powered by technology, enabled by pharmacology, informed by clinical science

While prevention is a bedrock of public health, innovative therapeutics are needed to complement the armamentarium of interventions required to achieve disease control and elimination targets for neglected diseases. Extraordinary advances in drug discovery technologies have occurred over the past decades, along with accumulation of scientific knowledge and experience in pharmacological and clinical sciences that are transforming many aspects of drug R&D across disciplines. We reflect on how these advances have propelled drug discovery for parasitic infections, focusing on malaria, kinetoplastid diseases, and cryptosporidiosis. We also discuss challenges and research priorities to accelerate discovery and development of urgently needed novel antiparasitic drugs.

Explainable Machine Learning for Property Predictions in Compound Optimization

The prediction of compound properties from chemical structure is a main task for machine learning (ML) in medicinal chemistry. ML is often applied to large data sets in applications such as compound screening, virtual library enumeration, or generative chemistry. Albeit desirable, a detailed understanding of ML model decisions is typically not required in these cases. By contrast, compound optimization efforts rely on small data sets to identify structural modifications leading to desired property profiles. In this situation, if ML is applied, one usually is reluctant to make decisions based on predictions that cannot be rationalized. Only few ML methods are interpretable. However, to yield insights into complex ML model decisions, explanatory approaches can be applied. Herein, methodologies for better understanding of ML models or explaining individual predictions are reviewed and current challenges in integrating ML into medicinal chemistry programs as well as future opportunities are discussed.

Active Site Sequence Representations of Human Kinases Outperform Full Sequence Representations for Affinity Prediction and Inhibitor Generation: 3D Effects in a 1D Model

Recent advances in deep learning have enabled the development of large-scale multimodal models for virtual screening and de novo molecular design. The human kinome with its abundant sequence and inhibitor data presents an attractive opportunity to develop proteochemometric models that exploit the size and internal diversity of this family of targets. Here, we challenge a standard practice in sequence-based affinity prediction models: instead of leveraging the full primary structure of proteins, each target is represented by a sequence of 29 discontiguous residues defining the ATP binding site. In kinase-ligand binding affinity prediction, our results show that the reduced active site sequence representation is not only computationally more efficient but consistently yields significantly higher performance than the full primary structure. This trend persists across different models, data sets, and performance metrics and holds true when predicting pIC₅₀ for both unseen ligands and kinases. Our interpretability analysis reveals a potential explanation for the superiority of the active site models: whereas only mild statistical effects about the extraction of three-dimensional (3D) interaction sites take place in the full sequence models, the active site models are equipped with an implicit but strong inductive bias about the 3D structure stemming from the discontiguity of the active sites. Moreover, in direct comparisons, our models perform similarly or better than previous state-of-the-art approaches in affinity prediction. We then investigate a de novo molecular design task and find that the active site provides benefits in the computational efficiency, but otherwise, both kinase representations yield similar optimized affinities (for both SMILES- and SELFIES-based molecular generators). Our work challenges the assumption that the full primary structure is indispensable for modeling human kinases.

Collaborative Profile-QSAR: A Natural Platform for Building Collaborative Models among Competing Companies

Massively multitask bioactivity models that transfer learning between thousands of assays have been shown to work dramatically better than separate models trained on each individual assay. In particular, the applicability domain for a given model can expand from compounds similar to those tested in that specific assay to those tested across the full complement of contributing assays. If many large companies would share their assay data and train models on the superset, predictions should be better than what each company can do alone. However, a company's compounds, targets, and activities are among their most guarded trade secrets. Strategies have been proposed to share just the individual collaborators' models, without exposing any of the training data. Profile-QSAR (pQSAR) is a two-level, multitask, stacked model. It uses profiles of level-1 predictions from single-task models for thousands of assays as compound descriptors for level-2 models. This work describes its simple and natural adaptation to safe collaboration by model sharing. Broad model sharing has not yet been implemented across multiple large companies, so there are numerous unanswered questions. Novartis was formed from several mergers and acquisitions. In principle, this should allow an internal simulation of model sharing. In practice, the lack of metadata about the origins of compounds and assays made this difficult. Nevertheless, we have attempted to simulate this process and propose some findings: multitask pQSAR is always an improvement over single-task models; collaborative multitask modeling did not improve predictions on internal compounds; collaboration did improve predictions for external compounds but far less than the purely internal multitask modeling for internal compounds; collaborative models for external compounds increasingly improve as overlap between compound collections increases; combining profiles from inside and outside the company is not best, with internal predictions better using only the inside profile and external using only the outside profile, but a consensus of models using all three profiles is best on external compounds and a good compromise on internal compounds. We anticipate similar results from other model-sharing approaches. Indeed, since collaborative pQSAR through model sharing is mathematically identical to pQSAR using actual shared data, we believe our conclusions should apply to collaborative modeling by any current method even including the unlikely scenario of directly sharing all chemical structures and assay data.

Evolution of Novartis' Small Molecule Screening Deck Design

This article summarizes the evolution of the screening deck at the Novartis Institutes for BioMedical Research (NIBR). Historically, the screening deck was an assembly of all available compounds. In 2015, we designed a first deck to facilitate access to diverse subsets with optimized properties. We allocated the compounds as plated subsets on a 2D grid with property based ranking in one dimension and increasing structural redundancy in the other. The learnings from the 2015 screening deck were applied to the design of a next generation in 2019. We found that using traditional leadlikeness criteria (mainly MW, clogP) reduces the hit rates of attractive chemical starting points in subset screening. Consequently, the 2019 deck relies on solubility and permeability to select preferred compounds. The 2019 design also uses NIBR's experimental assay data and inferred biological activity profiles in addition to structural diversity to define redundancy across the compound sets.

Molecular insight into the anion effect and free volume effect of CO2 solubility in multivalent ionic liquids

For many years, experimental and theoretical studies have investigated the solubility of CO2 in a variety of ionic liquids (ILs), but the overarching absorption mechanism is still unclear. Currently, two different factors are believed to dominate the absorption performance: (a) the fractional free volume (FFV) accessible for absorption; and (b) the nature of the CO2 interactions with the anion species. The FFV is often more influential than the specific choice of the anion, but neither mechanism provides a complete picture. Herein, we have attempted to decouple these mechanisms in order to provide a more definitive molecular-level perspective of CO2 absorption in IL solvents. We simulate a series of nine different multivalent ILs comprised of imidazolium cations and sulfonate/sulfonimide anions tethered to benzene rings, along with a comprehensive analysis of the CO2 absorption and underlying molecular-level features. We find that the CO2 solubility has a very strong, linear correlation with respect to FFV, but only when comparisons are constrained to a common anion species. The choice of anion results in a fundamental remapping of the correlation between CO2 solubility and FFV. Overall, the free volume effect dominates in the ILs with smaller FFV values, while the choice of anion becomes more important in the systems with larger FFVs. Our proposed mechanistic map is intended to provide a more consistent framework for guiding further IL design for gas absorption applications.

Convolutional architectures for virtual screening

Background

A Virtual Screening algorithm has to adapt to the different stages of this process. Early screening needs to ensure that all bioactive compounds are ranked in the first positions despite of the number of false positives, while a second screening round is aimed at increasing the prediction accuracy.

Results

A novel CNN architecture is presented to this aim, which predicts bioactivity of candidate compounds on CDK1 using a combination of molecular fingerprints as their vector representation, and has been trained suitably to achieve good results as regards both enrichment factor and accuracy in different screening modes (98.55% accuracy in active-only selection, and 98.88% in high precision discrimination).

Conclusion

The proposed architecture outperforms state-of-the-art ML approaches, and some interesting insights on molecular fingerprints are devised.

QSAR-derived affinity fingerprints (part 2): modeling performance for potency prediction

Affinity fingerprints report the activity of small molecules across a set of assays, and thus permit to gather information about the bioactivities of structurally dissimilar compounds, where models based on chemical structure alone are often limited, and model complex biological endpoints, such as human toxicity and in vitro cancer cell line sensitivity. Here, we propose to model in vitro compound activity using computationally predicted bioactivity profiles as compound descriptors. To this aim, we apply and validate a framework for the calculation of QSAR-derived affinity fingerprints (QAFFP) using a set of 1360 QSAR models generated using K_i, K_d, IC₅₀ and EC₅₀ data from ChEMBL database. QAFFP thus represent a method to encode and relate compounds on the basis of their similarity in bioactivity space. To benchmark the predictive power of QAFFP we assembled IC₅₀ data from ChEMBL database for 18 diverse cancer cell lines widely used in preclinical drug discovery, and 25 diverse protein target data sets. This study complements part 1 where the performance of QAFFP in similarity searching, scaffold hopping, and bioactivity classification is evaluated. Despite being inherently noisy, we show that using QAFFP as descriptors leads to errors in prediction on the test set in the ~ 0.65-0.95 pIC₅₀ units range, which are comparable to the estimated uncertainty of bioactivity data in ChEMBL (0.76-1.00 pIC₅₀ units). We find that the predictive power of QAFFP is slightly worse than that of Morgan2 fingerprints and 1D and 2D physicochemical descriptors, with an effect size in the 0.02-0.08 pIC₅₀ units range. Including QSAR models with low predictive power in the generation of QAFFP does not lead to improved predictive power. Given that the QSAR models we used to compute the QAFFP were selected on the basis of data availability alone, we anticipate better modeling results for QAFFP generated using more diverse and biologically meaningful targets. Data sets and Python code are publicly available at https://github.com/isidroc/QAFFP_regression .

Using Predicted Bioactivity Profiles to Improve Predictive Modeling

Predictive modeling is a cornerstone in early drug development. Using information for multiple domains or across prediction tasks has the potential to improve the performance of predictive modeling. However, aggregating data often leads to incomplete data matrices that might be limiting for modeling. In line with previous studies, we show that by generating predicted bioactivity profiles, and using these as additional features, prediction accuracy of biological endpoints can be improved. Using conformal prediction, a type of confidence predictor, we present a robust framework for the calculation of these profiles and the evaluation of their impact. We report on the outcomes from several approaches to generate the predicted profiles on 16 datasets in cytotoxicity and bioactivity and show that efficiency is improved the most when including the p-values from conformal prediction as bioactivity profiles.

Data-Driven Construction of Antitumor Agents with Controlled Polypharmacology

Controlling which particular members of a large protein family are targeted by a drug is key to achieving a desired therapeutic response. In this study, we report a rational data-driven strategy for achieving restricted polypharmacology in the design of antitumor agents selectively targeting the TYRO3, AXL, and MERTK (TAM) family tyrosine kinases. Our computational approach, based on the concept of fragments in structural environments (FRASE), distills relevant chemical information from structural and chemogenomic databases to assemble a three-dimensional inhibitor structure directly in the protein pocket. Target engagement by the inhibitors designed led to disruption of oncogenic phenotypes as demonstrated in enzymatic assays and in a panel of cancer cell lines, including acute lymphoblastic and myeloid leukemia (ALL/AML) and nonsmall cell lung cancer (NSCLC). Structural rationale underlying the approach was corroborated by X-ray crystallography. The lead compound demonstrated potent target inhibition in a pharmacodynamic study in leukemic mice.

All-Assay-Max2 pQSAR: Activity Predictions as Accurate as Four-Concentration IC50s for 8558 Novartis Assays

Profile-quantitative structure-activity relationship (pQSAR) is a massively multitask, two-step machine learning method with unprecedented scope, accuracy, and applicability domain. In step one, a "profile" of conventional single-assay random forest regression models are trained on a very large number of biochemical and cellular pIC₅₀ assays using Morgan 2 substructural fingerprints as compound descriptors. In step two, a panel of partial least squares (PLS) models are built using the profile of pIC₅₀ predictions from those random forest regression models as compound descriptors (hence the name). Previously described for a panel of 728 biochemical and cellular kinase assays, we have now built an enormous pQSAR from 11 805 diverse Novartis (NVS) IC₅₀ and EC₅₀ assays. This large number of assays, and hence of compound descriptors for PLS, dictated reducing the profile by only including random forest regression models whose predictions correlate with the assay being modeled. The random forest regression and pQSAR models were evaluated with our "realistically novel" held-out test set, whose median average similarity to the nearest training set member across the 11 805 assays was only 0.34, comparable to the novelty of compounds actually selected from virtual screens. For the 11 805 single-assay random forest regression models, the median correlation of prediction with the experiment was only r_ext² = 0.05, virtually random, and only 8% of the models achieved our standard success threshold of r_ext² = 0.30. For pQSAR, the median correlation was r_ext² = 0.53, comparable to four-concentration experimental IC₅₀s, and 72% of the models met our r_ext² > 0.30 standard, totaling 8558 successful models. The successful models included assays from all of the 51 annotated target subclasses, as well as 4196 phenotypic assays, indicating that pQSAR can be applied to virtually any disease area. Every month, all models are updated to include new measurements, and predictions are made for 5.5 million NVS compounds, totaling 50 billion predictions. Common uses have included virtual screening, selectivity design, toxicity and promiscuity prediction, mechanism-of-action prediction, and others. Several such actual applications are described.

Imputation of Assay Bioactivity Data Using Deep Learning

We describe a novel deep learning neural network method and its application to impute assay pIC₅₀ values. Unlike conventional machine learning approaches, this method is trained on sparse bioactivity data as input, typical of that found in public and commercial databases, enabling it to learn directly from correlations between activities measured in different assays. In two case studies on public domain data sets we show that the neural network method outperforms traditional quantitative structure-activity relationship (QSAR) models and other leading approaches. Furthermore, by focusing on only the most confident predictions the accuracy is increased to R² > 0.9 using our method, as compared to R² = 0.44 when reporting all predictions.

How to design potent and selective DYRK1B inhibitors? Molecular modeling study

DYRK1B protein kinase is an emerging anticancer target due to its overexpression in a variety of cancers and its role in cancer chemoresistance through maintaining cancer cells in the G0 (quiescent) state. Consequently, there is a growing interest in the development of potent and selective DYRK1B inhibitors for anticancer therapy. One of the major off-targets is another protein kinase, GSK3β, which phosphorylates an important regulator of cell cycle progression on the same residue as DYRK1B and is involved in multiple signaling pathways. In the current work, we performed a detailed comparative structural analysis of DYRK1B and GSK3β ATP-binding sites and identified key regions responsible for selectivity. As the crystal structure of DYRK1B has never been reported, we built and optimized a homology model by comparative modeling and metadynamics simulations. Calculation of interaction energies between docked ligands in the ATP-binding sites of both kinases allowed us to pinpoint key residues responsible for potency and selectivity. Specifically, the role of the gatekeeper residues in DYRK1B and GSK3β is discussed in detail, and two other residues are identified as key to selectivity of DYRK1B inhibition versus GSK3β. The analysis presented in this work was used to support the design of potent and selective azaindole-quinoline-based DYRK1B inhibitors and can facilitate development of more selective inhibitors for DYRK kinases.

Artificial Intelligence in Drug Design

Artificial Intelligence (AI) plays a pivotal role in drug discovery. In particular artificial neural networks such as deep neural networks or recurrent networks drive this area. Numerous applications in property or activity predictions like physicochemical and ADMET properties have recently appeared and underpin the strength of this technology in quantitative structure-property relationships (QSPR) or quantitative structure-activity relationships (QSAR). Artificial intelligence in de novo design drives the generation of meaningful new biologically active molecules towards desired properties. Several examples establish the strength of artificial intelligence in this field. Combination with synthesis planning and ease of synthesis is feasible and more and more automated drug discovery by computers is expected in the near future.

How to Achieve Better Results Using PASS-Based Virtual Screening: Case Study for Kinase Inhibitors

Discovery of new pharmaceutical substances is currently boosted by the possibility of utilization of the Synthetically Accessible Virtual Inventory (SAVI) library, which includes about 283 million molecules, each annotated with a proposed synthetic one-step route from commercially available starting materials. The SAVI database is well-suited for ligand-based methods of virtual screening to select molecules for experimental testing. In this study, we compare the performance of three approaches for the analysis of structure-activity relationships that differ in their criteria for selecting of "active" and "inactive" compounds included in the training sets. PASS (Prediction of Activity Spectra for Substances), which is based on a modified Naïve Bayes algorithm, was applied since it had been shown to be robust and to provide good predictions of many biological activities based on just the structural formula of a compound even if the information in the training set is incomplete. We used different subsets of kinase inhibitors for this case study because many data are currently available on this important class of drug-like molecules. Based on the subsets of kinase inhibitors extracted from the ChEMBL 20 database we performed the PASS training, and then applied the model to ChEMBL 23 compounds not yet present in ChEMBL 20 to identify novel kinase inhibitors. As one may expect, the best prediction accuracy was obtained if only the experimentally confirmed active and inactive compounds for distinct kinases in the training procedure were used. However, for some kinases, reasonable results were obtained even if we used merged training sets, in which we designated as inactives the compounds not tested against the particular kinase. Thus, depending on the availability of data for a particular biological activity, one may choose the first or the second approach for creating ligand-based computational tools to achieve the best possible results in virtual screening.

Protocols for the Design of Kinase-focused Compound Libraries

Protocols for the design of kinase-focused compound libraries are presented. Kinase-focused compound libraries can be differentiated based on the design goal. Depending on whether the library should be a discovery library specific for one particular kinase, a general discovery library for multiple distinct kinase projects, or even phenotypic screening, there exists today a variety of in silico methods to design candidate compound libraries. We address the following scenarios: 1) Datamining of SAR databases and kinase focused vendor catalogues; 2) Predictions and virtual screening; 3) Structure-based design of combinatorial kinase inhibitors; 4) Design of covalent kinase inhibitors; 5) Design of macrocyclic kinase inhibitors; and 6) Design of allosteric kinase inhibitors and activators.

Identification of Potent and Selective RIPK2 Inhibitors for the Treatment of Inflammatory Diseases

NOD2 (nucleotide-binding oligomerization domain-containing protein 2) is an internal pattern recognition receptor that recognizes bacterial peptidoglycan and stimulates host immune responses. Dysfunction of NOD2 pathway has been associated with a number of autoinflammatory disorders. To date, direct inhibitors of NOD2 have not been described due to technical challenges of targeting the oligomeric protein complex. Receptor interacting protein kinase 2 (RIPK2) is an intracellular serine/threonine/tyrosine kinase, a key signaling partner, and an obligate kinase for NOD2. As such, RIPK2 represents an attractive target to probe the pathological roles of NOD2 pathway. To search for selective RIPK2 inhibitors, we employed virtual library screening (VLS) and structure based design that eventually led to a potent and selective RIPK2 inhibitor 8 with excellent oral bioavailability, which was used to evaluate the effects of inhibition of RIPK2 in various in vitro assays and ex vivo and in vivo pharmacodynamic models.

Profile-QSAR 2.0: Kinase Virtual Screening Accuracy Comparable to Four-Concentration IC50s for Realistically Novel Compounds

While conventional random forest regression (RFR) virtual screening models appear to have excellent accuracy on random held-out test sets, they prove lacking in actual practice. Analysis of 18 historical virtual screens showed that random test sets are far more similar to their training sets than are the compounds project teams actually order. A new, cluster-based "realistic" training/test set split, which mirrors the chemical novelty of real-life virtual screens, recapitulates the poor predictive power of RFR models in real projects. The original Profile-QSAR (pQSAR) method greatly broadened the domain of applicability over conventional models by using as independent variables a profile of activity predictions from all historical assays in a large protein family. However, the accuracy still fell short of experiment on realistic test sets. The improved "pQSAR 2.0" method replaces probabilities of activity from naïve Bayes categorical models at several thresholds with predicted IC₅₀s from RFR models. Unexpectedly, the high accuracy also requires removing the RFR model for the actual assay of interest from the independent variable profile. With these improvements, pQSAR 2.0 activity predictions are now statistically comparable to medium-throughput four-concentration IC₅₀ measurements even on the realistic test set. Beyond the yes/no activity predictions from a typical high-throughput screen (HTS) or conventional virtual screen, these semiquantitative IC₅₀ predictions allow for predicted potency, ligand efficiency, lipophilic efficiency, and selectivity against antitargets, greatly facilitating hitlist triaging and enabling virtual screening panels such as toxicity panels and overall promiscuity predictions.

Prediction of Protein Kinase-Ligand Interactions through 2.5D Kinochemometrics

So far, 518 protein kinases have been identified in the human genome. They share a common mechanism of protein phosphorylation and are involved in many critical biological processes of eukaryotic cells. Deregulation of the kinase phosphorylation function induces severe illnesses such as cancer, diabetes, or inflammatory diseases. Many actors in the pharmaceutical domain have made significant efforts to design potent and selective protein kinase inhibitors as new potential drugs. Because the ATP binding site is highly conserved in the protein kinase family, the design of selective inhibitors remains a challenge and has negatively impacted the progression of drug candidates to late-stage clinical development. The work presented here adopts a 2.5D kinochemometrics (KCM) approach, derived from proteochemometrics (PCM), in which protein kinases are depicted by a novel 3D descriptor and the ligands by 2D fingerprints. We demonstrate in two examples that the protein descriptor successfully classified protein kinases based on their group membership and their Asp-Phe-Gly (DFG) conformation. We also compared the performance of our models with those obtained from a full 2D KCM model and QSAR models. In both cases, the internal validation of the models demonstrated good capabilities to distinguish "active" from "inactive" protein kinase-ligand pairs. However, the external validation performed on two independent data sets showed that the two statistical models tended to overestimate the number of "inactive" pairs.

Profiling Prediction of Kinase Inhibitors: Toward the Virtual Assay

Kinome-wide screening would have the advantage of providing structure-activity relationships against hundreds of targets simultaneously. Here, we report the generation of ligand-based activity prediction models for over 280 kinases by employing Machine Learning methods on an extensive data set of proprietary bioactivity data combined with open data. High quality (AUC > 0.7) was achieved for ∼200 kinases by (1) combining open with proprietary data, (2) choosing Random Forest over alternative tested Machine Learning methods, and (3) balancing the training data sets. Tests on left-out and external data indicate a high value for virtual screening projects. Importantly, the derived models are evenly distributed across the kinome tree, allowing reliable profiling prediction for all kinase branches. The prediction quality was further improved by employing experimental bioactivity fingerprints of a small kinase subset. Overall, the generated models can support various hit identification tasks, including virtual screening, compound repurposing, and the detection of potential off-targets.

Toward the Validation of Maternal Embryonic Leucine Zipper Kinase: Discovery, Optimization of Highly Potent and Selective Inhibitors, and Preliminary Biology Insight

MELK kinase has been implicated in playing an important role in tumorigenesis. Our previous studies suggested that MELK is involved in the regulation of cell cycle and its genetic depletion leads to growth inhibition in a subset of high MELK-expressing basal-like breast cancer cell lines. Herein we describe the discovery and optimization of novel MELK inhibitors 8a and 8b that recapitulate the cellular effects observed by short hairpin ribonucleic acid (shRNA)-mediated MELK knockdown in cellular models. We also discovered a novel fluorine-induced hydrophobic collapse that locked the ligand in its bioactive conformation and led to a 20-fold gain in potency. These novel pharmacological inhibitors achieved high exposure in vivo and were well tolerated, which may allow further in vivo evaluation.

An overview of molecular fingerprint similarity search in virtual screening

INTRODUCTION

A central premise of medicinal chemistry is that structurally similar molecules exhibit similar biological activities. Molecular fingerprints encode properties of small molecules and assess their similarities computationally through bit string comparisons. Based on the similarity to a biologically active template, molecular fingerprint methods allow for identifying additional compounds with a higher chance of displaying similar biological activities against the same target - a process commonly referred to as virtual screening (VS).

AREAS COVERED

This article focuses on fingerprint similarity searches in the context of compound selection for enhancing hit sets, comparing compound decks, and VS. In addition, the authors discuss the application of fingerprints in predictive modeling.

EXPERT OPINION

Fingerprint similarity search methods are especially useful in VS if only a few unrelated ligands are known for a given target and therefore more complex and information rich methods such as pharmacophore searches or structure-based design are not applicable. In addition, fingerprint methods are used in characterizing properties of compound collections such as chemical diversity, density in chemical space, and content of biologically active molecules (biodiversity). Such assessments are important for deciding what compounds to experimentally screen, to purchase, or to assemble in a virtual compound deck for in silico screening or de novo design.

Perspective on computational and structural aspects of kinase discovery from IPK2014

Recent advances in understanding the activity and selectivity of kinase inhibitors and their relationships to protein structure are presented. Conformational selection in kinases is studied from empirical, data-driven and simulation approaches. Ligand binding and its affinity are, in many cases, determined by the predetermined active and inactive conformation of kinases. Binding affinity and selectivity predictions highlight the current state of the art and advances in computational chemistry as it applies to kinase inhibitor discovery. Kinome wide inhibitor profiling and cell panel profiling lead to a better understanding of selectivity and allow for target validation and patient tailoring hypotheses. This article is part of a Special Issue entitled: Inhibitors of Protein Kinases.

Discovery of novel drugs for promising targets

BACKGROUND

Once a promising drug target is identified, the steps to actually discover and optimize a drug are diverse and challenging.

OBJECTIVE

The goal of this study was to provide a road map to navigate drug discovery.

METHODS

Review general steps for drug discovery and provide illustrating references.

RESULTS

A number of approaches are available to enhance and accelerate target identification and validation. Consideration of a variety of potential mechanisms of action of potential drugs can guide discovery efforts. The hit to lead stage may involve techniques such as high-throughput screening, fragment-based screening, and structure-based design, with informatics playing an ever-increasing role. Biologically relevant screening models are discussed, including cell lines, 3-dimensional culture, and in vivo screening. The process of enabling human studies for an investigational drug is also discussed.

CONCLUSIONS

Drug discovery is a complex process that has significantly evolved in recent years.

Predictive in silico off-target profiling in drug discovery

Drug action can be rationalized as interaction of a molecule with proteins in a regulatory network of targets from a specific biological system. Both drug and side effects are often governed by interaction of the drug molecule with many, often unrelated, targets. Accordingly, arrays of protein–ligand interaction data from numerous in vitro profiling assays today provide growing evidence of polypharmacological drug interactions, even for marketed drugs. In vitro off-target profiling has therefore become an important tool in early drug discovery to learn about potential off-target liabilities, which are sometimes beneficial, but more often safety relevant. The rapidly developing field of in silico profiling approaches is complementing in vitro profiling. These approaches capitalize from large amounts of biochemical data from multiple sources to be exploited for optimizing undesirable side effects in pharmaceutical research. Therefore, current in silico profiling models are nowadays perceived as valuable tools in drug discovery, and promise a platform to support optimally informed decisions.

Cheminformatics aspects of high throughput screening: from robots to models: symposium summary

The "Cheminformatics aspects of high throughput screening (HTS): from robots to models" symposium was part of the computers in chemistry technical program at the American Chemical Society National Meeting in Denver, Colorado during the fall of 2011. This symposium brought together researchers from high throughput screening centers and molecular modelers from academia and industry to discuss the integration of currently available high throughput screening data and assays with computational analysis. The topics discussed at this symposium covered the data-infrastructure at various academic, hospital, and National Institutes of Health-funded high throughput screening centers, the cheminformatics and molecular modeling methods used in real world examples to guide screening and hit-finding, and how academic and non-profit organizations can benefit from current high throughput screening cheminformatics resources. Specifically, this article also covers the remarks and discussions in the open panel discussion of the symposium and summarizes the following talks on "Accurate Kinase virtual screening: biochemical, cellular and selectivity", "Selective, privileged and promiscuous chemical patterns in high-throughput screening" and "Visualizing and exploring relationships among HTS hits using network graphs".

Large-Scale QSAR in Target Prediction and Phenotypic HTS Assessment

The advent of in silico compound target prediction offers a potential paradigm shift in how large compound collections are understood and used strategically in high-throughput screens (HTS). Specifically, phenotypic HTS hits may be annotated both with known targets and predicted targets using large-scale QSAR models, enabling a more sophisticated hit assessment. Efforts in massive bioactivity data integration and standardization is empowering such compound-target annotations. These approaches differ fundamentally from the traditional role of QSAR in lead optimization and binding affinity predictions to global, probabilistic target predictions for thousands of human proteins.

Structural approaches to obtain kinase selectivity

One of the grand challenges in kinase drug discovery is the design of small-molecule inhibitors with selectivity profiles that will ultimately be efficacious in the clinic. Current medicinal chemistry strategies make heavy use of structural, biophysical and computational approaches to achieve this multi-faceted goal. Here we review structure-based approaches underlying the development of several molecules that are currently in clinical trials, including the cMet inhibitor ARQ197 and the Bcr-Abl inhibitor ponatinib. We highlight the challenge posed by the emergence of resistance mutants and discuss promising lead generation strategies to obtain selective inhibitors of protein and lipid kinases such as targeting of specific sites, the use of fragment-based approaches and new chemical probes based on metal complexes.