Prior to clinical development, a comprehensive pharmacokinetic characterization of a novel drug is required to understand its exposure at the site of action and elimination. Accordingly, in vitro assays and animal pharmacokinetic studies are regularly employed to predict drug exposure in humans, which is often costly and time-consuming. For this reason, the prediction of human pharmacokinetics at the point of design would be of high value for drug discovery. Therefore, we have established a comprehensive data curation protocol that enables machine learning evaluation of 12 human in vivo pharmacokinetic parameters using only chemical structure information and available doses for 1001 unique compounds. These machine learning models were thoroughly investigated and validated using both an independent hold-out test set and AstraZeneca clinical data. In addition, the availability of preclinical predictions for a subset of internal clinical candidates allowed us to compare our in silico approach with state-of-the-art pharmacokinetic predictions. Based on this evaluation, three fit-for-purpose models for AUC PO (R test 2 = 0.63; RMSEtest = 0.76), C max PO (R test 2 = 0.68; RMSEtest = 0.62), and Vdss IV (R test 2 = 0.47; RMSEtest = 0.50) were identified. Based on the findings, our machine learning models have considerable potential for practical applications in drug discovery, such as influencing decision-making in drug discovery projects and progression of drug candidates toward the clinic.
Noncovalent inhibitors of protein kinases have different modes of action. They bind to the active or inactive form of kinases, compete with ATP, stabilize inactive kinase conformations, or act through allosteric sites. Accordingly, kinase inhibitors have been classified on the basis of different binding modes. For medicinal chemistry, it would be very useful to derive mechanistic hypotheses for newly discovered inhibitors. Therefore, we have applied different machine learning approaches to generate models for predicting different classes of kinase inhibitors including types I, I 1 / 2 , and II as well as allosteric inhibitors. These models were built on the basis of compounds with binding modes confirmed by X-ray crystallography and yielded unexpectedly accurate and stable predictions without the need for deep learning. The results indicate that the new machine learning models have considerable potential for practical applications. Therefore, our data sets and models are made freely available.
Allosteric kinase inhibitors are thought to have high selectivity and are prime candidates for kinase drug discovery. In addition, the exploration of allosteric mechanisms represents an attractive topic for basic research and drug design. Although the identification and characterization of allosteric kinase inhibitors is still far from being routine, X-ray structures of kinase complexes have been determined for a significant number of such inhibitors. On the basis of structural data, allosteric inhibitors can be confirmed. We report a comprehensive survey of allosteric kinase inhibitors and activators from publicly available X-ray structures, map their binding sites, and determine their distribution over binding pockets in kinases. In addition, we discuss structural features of these compounds and identify active structural analogues and highconfidence target annotations, indicating additional activities for a subset of allosteric inhibitors. This contribution aims to provide a detailed structure-based view of allosteric kinase inhibition.
Kinase inhibitors are high-priority drug candidates for a variety of therapeutic applications. Accordingly, there has been a rapid growth in the number of kinase inhibitors and volumes of associated activity data. A paradigm for the use of kinase inhibitors in oncology is that these compounds have multitarget activities and elicit their therapeutic effects through polypharmacology. An analysis of kinase inhibitors and associated activity data from medicinal chemistry has so far only identified small subsets of highly promiscuous kinase inhibitors. In this study, we have collected inhibitors of human kinases and their activity data from seven public repositories, curated, and combined these data, yielding more than 112 000 inhibitors with well-defined activity measurements from which qualitative target annotations were derived. An analysis of these unprecedentedly large data sets revealed that nearly 40% of human kinase inhibitors have multikinase activities but that only 4% are known to be active against five or more kinases. However, structurally analogous inhibitors often displayed significant differences in the number of kinase annotations, leading to the formation of nearly 16 000 “promiscuity cliffs”. Moreover, 2236 promiscuity cliffs (14.03%) were formed by kinase inhibitors at different stages of clinical development. Overall, these cliffs suggested many target hypotheses for kinase inhibitors, taking data incompleteness into consideration, as well as hypotheses for structural modifications leading to kinase selectivity. Furthermore, from network representations, pathways comprising sequences of promiscuity cliffs were extracted that revealed unexpected structure–promiscuity relationships. To enable follow-up investigations, all promiscuity cliffs formed by human kinase inhibitors will be made freely available.
Predicting drug-target interaction is key for drug discovery. Recent deep learning-based methods show promising performance but two challenges remain: (i) how to explicitly model and learn local interactions between drugs and targets for better prediction and interpretation; (ii) how to generalize prediction performance on novel drug-target pairs from different distribution. In this work, we propose DrugBAN, a deep bilinear attention network (BAN) framework with domain adaptation to explicitly learn pair-wise local interactions between drugs and targets, and adapt on out-of-distribution data. DrugBAN works on drug molecular graphs and target protein sequences to perform prediction, with conditional domain adversarial learning to align learned interaction representations across different distributions for better generalization on novel drug-target pairs. Experiments on three benchmark datasets under both in-domain and cross-domain settings show that DrugBAN achieves the best overall performance against five state-of-the-art baselines. Moreover, visualizing the learned bilinear attention map provides interpretable insights from prediction results. D rug-target interaction (DTI) prediction serves as an important step in the process of drug discovery 1-3 . Traditional biomedical measuring from in vitro experiments is reliable but has notably high cost and time-consuming development cycle, preventing its application on large-scale data 4 . In contrast, identifying high-confidence DTI pairs by in silico approaches can greatly narrow down the search scope of compound candidates, and provide insights into the causes of potential side effects in drug combinations. Therefore, in silico approaches have gained increasing attention and made much progress in the last few years 5, 6 .For in silico approaches, traditional structure-based and ligand-based virtual screening (VS) methods have been studied widely for their decent performance 7 . However, structure-based VS requires molecular docking simulation, which is not applicable if the target protein's three-dimensional (3D) structure is unknown. On the other hand, ligand-based VS predicts new active molecules based on the known actives to the same protein, but the performance is poor when the number of known actives is insufficient 8 . More recently, deep learning (DL)-based approaches have rapidly progressed for computational DTI prediction due to their successes in other areas, enabling large-scale validation in a relatively short time 9 . Many of them are constructed from a chemogenomics perspective 3,10 , which integrates the chemical space, genomic space, and interaction information into a unified end-to-end framework. Since the number of biological targets that have available 3D structures is limited, many DL-based models take linear or two-dimensional (2D) structural information of drugs and proteins as inputs. They treat DTI prediction as a binary classification task, and make predictions by feeding the inputs into different deep encoding and decoding modules such as...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.