RESUMO
We present a novel computational approach for predicting human pharmacokinetics (PK) that addresses the challenges of early stage drug design. Our study introduces and describes a large-scale data set of 11 clinical PK end points, encompassing over 2700 unique chemical structures to train machine learning models. To that end multiple advanced training strategies are compared, including the integration of in vitro data and a novel self-supervised pretraining task. In addition to the predictions, our final model provides meaningful epistemic uncertainties for every data point. This allows us to successfully identify regions of exceptional predictive performance, with an absolute average fold error (AAFE/geometric mean fold error) of less than 2.5 across multiple end points. Together, these advancements represent a significant leap toward actionable PK predictions, which can be utilized early on in the drug design process to expedite development and reduce reliance on nonclinical studies.
RESUMO
Although computational predictions of pharmacokinetics (PK) are desirable at the drug design stage, existing approaches are often limited by prediction accuracy and human interpretability. Using a discovery data set of mouse and rat PK studies at Roche (9,685 unique compounds), we performed a proof-of-concept study to predict key PK properties from chemical structure alone, including plasma clearance (CLp), volume of distribution at steady-state (Vss), and oral bioavailability (F). Ten machine learning (ML) models were evaluated, including Single-Task, Multitask, and transfer learning approaches (i.e., pretraining with in vitro data). In addition to prediction accuracy, we emphasized human interpretability of outcomes, especially the quantification of uncertainty, applicability domains, and explanations of predictions in terms of molecular features. Results show that intravenous (IV) PK properties (CLp and Vss) can be predicted with good precision (average absolute fold error, AAFE of 1.96-2.84 depending on data split) and low bias (average fold error, AFE of 0.98-1.36), with AutoGluon, Gaussian Process Regressor (GP), and ChemProp displaying the best performance. Driven by higher complexity of oral PK studies, predictions of F were more challenging, with the best AAFE values of 2.35-2.60 and higher overprediction bias (AFE of 1.45-1.62). Multi-Task approaches and pretraining of ChemProp neural networks with in vitro data showed similar precision to Single-Task models but helped reduce the bias and increase correlations between observations and predictions. A combination of GP-computed prediction variance, molecular clustering, and dimensionality-reduction provided valuable quantitative insights into prediction uncertainty and applicability domains. SHAPley Additive exPlanations (SHAPs) highlighted molecular features contributing to prediction outcomes of Vss, providing explanations that could aid drug design. Combined results show that computational predictions of PK are feasible at the drug design stage, with several ML technologies converging to successfully leverage historical PK data sets. Further studies are needed to unlock the full potential of this approach, especially with respect to data set sizes and quality, transfer learning between in vitro and in vivo data sets, model-independent quantification of uncertainty, and explainability of predictions.
Assuntos
Desenho de Fármacos , Redes Neurais de Computação , Humanos , Ratos , Animais , Disponibilidade Biológica , Administração Intravenosa , Farmacocinética , Modelos Biológicos , Preparações FarmacêuticasRESUMO
The keeping of backyard poultry and waterfowl as pets has become increasingly popular in recent years, resulting in a rising case load of these patients in veterinary practices. Diagnostic imaging techniques are taking a leading role in rapid diagnosis in the live bird. We provide an overview of the most important points regarding radiographic and ultrasonographic imaging procedures in these birds. We also review the most commonly documented radiographic and ultrasonographic signs in these species, as well as discuss unique anatomic characteristics with which the veterinarian should be familiar.
Assuntos
Anseriformes , Doenças das Aves/diagnóstico por imagem , Aves Domésticas , Radiografia/veterinária , Ultrassonografia/veterinária , Animais , Doenças das Aves Domésticas/diagnóstico por imagem , Radiografia/métodos , Ultrassonografia/métodosRESUMO
Pd-catalyzed C-N couplings are commonplace in academia and industry. Despite their significance, finding suitable reaction conditions leading to a high yield, for instance, remains a challenging and time-consuming task which usually requires screening over many sets of conditions. To help select promising reaction conditions in the vast space of reagent combinations, machine learning is an emerging technique with a lot of promise. In this work, we assess whether the reaction yield of C-N couplings can be predicted from databases of chemical reactions. We test the generalizability of models both on challenging data splits and on a dedicated experimental test set. We find that, provided the chemical space represented by the training set is not left, the models perform well. However, the applicability domain is quickly left even for simple reactions of the same type, as, for instance, present in our plate test set. The results show that yield prediction for new reactions is possible from the algorithmic side but in practice is hindered by the available data. Most importantly, more data that cover the diversity in reagents are needed for a general-purpose prediction of reaction yields. Our findings also expose a challenge to this field in that it appears to be extremely deceiving to judge models based on literature data with test sets which are split off the same literature data, even when challenging splits are considered.
RESUMO
The G-protein-coupled cannabinoid receptor type 2 (CB2R) is a key element of the endocannabinoid (EC) system. EC/CB2R signaling has significant therapeutic potential in major pathologies affecting humans such as allergies, neurodegenerative disorders, inflammation or ocular diseases. CB2R agonism exerts anti-inflammatory and tissue protective effects in preclinical animal models of cardiovascular, gastrointestinal, liver, kidney, lung and neurodegenerative disorders. Existing ligands can be subdivided into endocannabinoids, cannabinoid-like and synthetic CB2R ligands that possess various degrees of potency on and selectivity against the cannabinoid receptor type 1. This review is an account of granted CB2R ligand patents from 2010 up to the present, which were surveyed using Derwent Innovation®.
Assuntos
Anti-Inflamatórios , Endocanabinoides , Animais , Humanos , Ligantes , Patentes como Assunto , Receptores de Canabinoides , Transdução de SinaisRESUMO
Despite the widespread and increasing usage of Pd-catalyzed C-N cross couplings, finding good conditions for these reactions can be challenging. Practitioners mostly rely on few methodology studies or anecdotal experience. This is surprising, since the advent of data-driven experimentation and the large amount of knowledge in databases allow for data-driven insight. In this work, we address this by analyzing more than 62 000 Buchwald-Hartwig couplings gathered from CAS, Reaxys and the USPTO. Our meta-analysis of the reaction performance generates data-driven cheatsheets for reaction condition recommendation. It also provides an interactive tool to find rarer ligands with optimal performance regarding user-selected substrate properties. With this we give practitioners promising starting points. Furthermore, we study bias and diversity in the literature and summarize the current state of the reaction data, including its pitfalls. Hence, this work will also be useful for future data-driven developments such as the optimization of reaction conditions via machine learning.
RESUMO
Widely used regression approaches in modeling quantitative structure-property relationships, such as PLS regression, are highly susceptible to outlying observations that will impair the prognostic value of a model. Our aim is to compile homogeneous datasets as the basis for regression modeling by removing outlying compounds and applying variable selection. We investigate different approaches to create robust, outlier-resistant regression models in the field of prediction of drug molecules' permeability. The objective is to join the strength of outlier detection and variable elimination increasing the predictive power of prognostic regression models. In conclusion, outlier detection is employed to identify multiple, homogeneous data subsets for regression modeling.
Assuntos
Modelos Biológicos , Modelos Químicos , Preparações Farmacêuticas/química , Relação Quantitativa Estrutura-Atividade , Análise de Regressão , Algoritmos , Humanos , Permeabilidade , FarmacologiaRESUMO
The first large scale analysis of in vitro absorption, distribution, metabolism, excretion, and toxicity (ADMET) data shared across multiple major pharma has been performed. Using advanced matched molecular pair analysis (MMPA), we combined data from three pharmaceutical companies and generated ADMET rules, avoiding the need to disclose the full chemical structures. On top of the very large exchange of knowledge, all companies involved synergistically gained approximately 20% more rules from the shared transformations. There is good quantitative agreement between the rules based on shared data compared to both individual companies' rules and rules published in the literature. Known correlations between log D, solubility, in vitro clearance, and plasma protein binding also hold in transformation space, but there are also interesting exceptions. Data pools such as this allow focusing on particular functional groups and characterizing their ADMET profile. Finally the role of a corpus of robustly tested medicinal chemistry knowledge in the training of medicinal chemistry is discussed.
Assuntos
Química Farmacêutica/estatística & dados numéricos , Indústria Farmacêutica/estatística & dados numéricos , Farmacologia/métodos , Animais , Mineração de Dados , Conjuntos de Dados como Assunto , Cães , Humanos , Macaca fascicularis , Células Madin Darby de Rim Canino , Taxa de Depuração Metabólica , Camundongos , Farmacologia/estatística & dados numéricos , Ligação Proteica , Ratos , SolubilidadeRESUMO
We have investigated techniques for distinguishing between drugs and nondrugs using a set of molecular descriptors derived from semiempirical molecular orbital (AM1) calculations. The "drug" data set of 2105 compounds was derived from the World Drug Index (WDI) using a procedure designed to select real drugs. The "nondrug" data set was the Maybridge database. We have first investigated the dimensionality of physical properties space based on a set of 26 descriptors that we have used successfully to build absorption, distribution, metabolism, and excretion-related quantitative structure-property relationship models. We discuss the general nature of the descriptors for physical property space and the ability of these descriptors to distinguish between drugs and nondrugs. The third most significant principal component of this set of descriptors serves as a useful numerical index of drug-likeness, but no others are able to distinguish between drugs and nondrugs. We have therefore extended our set of descriptors to a total of 66 and have used recursive partitioning to identify the descriptors that can distinguish between drugs and nondrugs. This procedure pointed to two of the descriptors that play an important role in the principal component found above and one more from the set of 40 extra descriptors. These three descriptors were then used to train a Kohonen artificial neural net for the entire Maybridge data set. Projecting the drug database onto the map obtained resulted in a clear distinction not only between drugs and nondrugs but also, for instance, between hormones and other drugs. Projection of 42 131 compounds from the WDI onto the Kohonen map also revealed pronounced clustering in the regions of the map assigned as druglike.
Assuntos
Preparações Farmacêuticas/química , Preparações Farmacêuticas/classificação , Fenômenos Químicos , Físico-Química , Bases de Dados Factuais , Desenho de Fármacos , Redes Neurais de Computação , Teoria QuânticaRESUMO
Posttranscriptional regulation and RNA metabolism have become central topics in the understanding of mammalian gene expression and cell signalling, with the 3' untranslated region emerging as the coordinating unit. The 3' untranslated region trans-acting factor Hu protein R (HuR) forms a central posttranscriptional pathway node bridging between AU-rich element-mediated processes and microRNA regulation. While (m)RNA control by HuR has been extensively characterized, the molecular mode of action still remains elusive. Here we describe the identification of the first RRM3 (RNA recognition motif 3) targeted low molecular weight HuR inhibitors from a one-bead-one-compound library screen using confocal nanoscanning. A further compound characterization revealed the presence of an ATP-binding pocket within HuR RRM3, associated with enzymatic activity. Centered around a metal-ion-coordinating DxD motif, the catalytic site mediates 3'-terminal adenosyl modification of non-polyadenylated RNA substrates by HuR. These findings suggest that HuR actively contributes to RNA modification and maturation and thereby shed an entirely new light on the role of HuR in RNA metabolism.