DruMAP: A Novel Drug Metabolism and Pharmacokinetics Analysis Platform

kMoL: an open-source machine and federated learning library for drug discovery

Machine learning is quickly becoming integral to drug discovery pipelines, particularly quantitative structure-activity relationship (QSAR) and absorption, distribution, metabolism, and excretion (ADME) tasks. Graph Convolutional Network (GCN) models have proven especially promising due to their inherent ability to model molecular structures using graph-based representations. However, maximizing the potential of such models in practice is challenging, as companies prioritize data privacy and security over collaboration initiatives to improve model performance and robustness. kMoL is an open-source machine learning library with integrated federated learning capabilities developed to address such challenges. Its key features include state-of-the-art model architectures, Bayesian optimization, explainability, and federated learning mechanisms. It demonstrates extensive customization possibilities, advanced security features, straightforward implementation of user-specific models, and high adaptability to custom datasets without additional programming requirements. kMoL is evaluated through locally trained benchmark settings and distributed federated learning experiments using various datasets to assess the features and flexibility of the library, as well as the ability to facilitate fast and practical experimentation. Additionally, results of these experiments provide further insights into the performance trade-offs associated with federated learning strategies, presenting valuable guidance for deploying machine learning models in a privacy-preserving manner within drug discovery pipelines. kMoL is available on GitHub at https://github.com/elix-tech/kmol .Scientific contribution The primary scientific contribution of this research project is the introduction and evaluation of kMoL, an open-source machine learning library with integrated federated learning capabilities. By demonstrating advanced customization and security capabilities without additional programming requirements, kMoL represents an accessible yet secure open-source platform for collaborative drug discovery projects. Additionally, the experiment results provide further insights into the performance trade-offs associated with federated learning strategies, presenting valuable guidance for deploying machine learning models in a privacy-preserving manner within drug discovery pipelines.

The state-of-the-art machine learning model for plasma protein binding prediction: Computational modeling with OCHEM and experimental validation

Plasma protein binding (PPB) is closely related to pharmacokinetics, pharmacodynamics and drug toxicity. Existing models for predicting PPB often suffer from low prediction accuracy and poor interpretability, especially for high PPB compounds, and are most often not experimentally validated. Here, we carried out a strict data curation protocol, and applied consensus modeling to obtain a model with a coefficient of determination of 0.90 and 0.91 on the training set and the test set, respectively. This model (available on the OCHEM platform https://ochem.eu/article/29) was further retrospectively validated for a set of 63 poly-fluorinated molecules and prospectively validated for a set of 25 highly diverse compounds, and its performance for both these sets was superior to that of the other previously reported models. Furthermore, we identified the physicochemical and structural characteristics of high and low PPB molecules for further structural optimization. Finally, we provide practical and detailed recommendations for structural optimization to decrease PPB binding of lead compounds.

Predicting the effects of drugs and unveiling their mechanisms of action using an interpretable pharmacodynamic mechanism knowledge graph (IPM-KG)

BACKGROUND

Multiple studies have aimed to consolidate drug-related data and predict drug effects. However, most of these studies have focused on integrating diverse data through correlation rather than representing them based on the pharmacodynamic mechanism of action (MOA). It is thus crucial to obtain interpretability to validate prediction results. In this study, we propose a novel framework to construct knowledge graphs that represent pharmacodynamic MOA, predict drug effects, and derive conceivable mechanistic pathways.

METHODS AND RESULTS

We constructed an interpretable pharmacodynamic mechanism knowledge graph (IPM-KG) by integrating various existing databases and combining them with the approach of this study to automatically fill in the missing data. This yielded a knowledge graph comprising 1455 drugs and 2547 diseases. Additionally, a graph neural network (GNN)-based approach was used to predict therapeutic medication and indication, which outperformed previous approaches that relied on correlation-based knowledge graphs lacking pharmacodynamic MOA representations. Furthermore, we proposed and assessed a method to interpret pharmacodynamic MOA using gene perturbation data. This feasibility study demonstrated the successful derivation of an accurate mechanism in approximately 50 % of cases. Additionally, it facilitated the identification of candidate drugs, which are currently unapproved but exhibit potential for drug repositioning, and their mechanisms of action.

CONCLUSIONS

This framework not only enables the derivation of highly accurate "drug-indication" predictions but also offers a basic mechanistic understanding, thereby facilitating future drug repositioning efforts.

Prediction of human pharmacokinetic parameters incorporating SMILES information

This study aimed to develop a model incorporating natural language processing analysis for the simplified molecular-input line-entry system (SMILES) to predict clearance (CL) and volume of distribution at steady state (Vd,ss) in humans. The construction of CL and Vd,ss prediction models involved data from 435 to 439 compounds, respectively. In machine learning, features such as animal pharmacokinetic data, in vitro experimental data, molecular descriptors, and SMILES were utilized, with XGBoost employed as the algorithm. The ChemBERTa model was used to analyze substance SMILES, and the last hidden layer embedding of ChemBERTa was examined as a feature. The model was evaluated using geometric mean fold error (GMFE), r2, root mean squared error (RMSE), and accuracy within 2- and 3-fold error. The model demonstrated optimal performance for CL prediction when incorporating animal pharmacokinetic data, in vitro experimental data, and SMILES as features, yielding a GMFE of 1.768, an r2 of 0.528, an RMSE of 0.788, with accuracies within 2-fold and 3-fold error reaching 75.8% and 81.8%, respectively. The model's performance in Vd,ss prediction was optimized by leveraging animal pharmacokinetic data and in vitro experimental data as features, yielding a GMFE of 1.401, an r2 of 0.902, an RMSE of 0.413, with accuracies within 2-fold and 3-fold error reaching 93.8% and 100%, respectively. This study has developed a highly predictive model for CL and Vd,ss. Specifically, incorporating SMILES information into the model has predictive power for CL.

AI-Enabled Ultra-large Virtual Screening Identifies Potential Inhibitors of Choline Acetyltransferase for Theranostic Purposes

Alzheimer's disease (AD) and related dementias are among the primary neurological disorders and call for the urgent need for early-stage diagnosis to gain an upper edge in therapeutic intervention and increase the overall success rate. Choline acetyltransferase (ChAT) is the key acetylcholine (ACh) biosynthesizing enzyme and a legitimate target for the development of biomarkers for early-stage diagnosis and monitoring of therapeutic responses. It is also a theranostic target for tackling colon and lung cancers, where overexpression of non-neuronal ChAT leads to the production of acetylcholine, which acts as an autocrine growth factor for cancer cells. Theranostics is a hybrid of diagnostics and therapeutics that can be used to locate cancer cells using radiotracers and kill them without affecting other healthy tissues. Traditional virtual screening protocols have a lot of limitations; given the current rate of chemical database expansion exceeding billions, much faster screening protocols are required. Deep docking (DD) is one such platform that leverages the power of deep neural network (DNN)-based virtual screening, empowering researchers to dock billions of molecules in a speedy, yet explicit manner. Here, we have screened 1.3 billion compounds library from the ZINC20 database, identifying the best-performing hits. With each iteration run where the first iteration gave ∼116 million hits, the second iteration gave ∼3.7 million hits, and the final third iteration gave 168,447 hits from which further refinement gave us the top 5 compounds as potential ChAT inhibitors. The discovery of novel ChAT inhibitors will enable researchers to develop new probes that can be used as novel theranostic agents against cancer and as early-stage diagnostics for the onset of AD, for timely therapeutic intervention to halt the further progression of AD.

Machine Learning ADME Models in Practice: Four Guidelines from a Successful Lead Optimization Case Study

Optimization of the ADME properties and pharmacokinetic (PK) profile of compounds is one of the critical activities in any medicinal chemistry campaign to discover a future clinical candidate. Finding ways to expedite the process to address ADME/PK shortcomings and reduce the number of compounds to synthesize is highly valuable. This article provides practical guidelines and a case study on the use of ML ADME models to guide compound design in small molecule lead optimization. These guidelines highlight that ML models cannot have an impact in a vacuum: they help advance a program when they have the trust of users, are tuned to the needs of the program, and are integrated into decision-making processes in a way that complements and augments the expertise of chemists.

Effect of chronic administration of ostruthin on depression-like behavior in chronically stressed mice

We have previously shown that a single dose of a TREK-1 channel activator, ostruthin, exhibited antidepressant and anxiolytic effects in acute behavioral test models in mice. To assess the potential clinical application, it is essential to evaluate the effects of long-term administration of ostruthin in a chronically stressed mouse model, which is considered to be similar to the clinical condition of major depression in humans. Here, we tested the effects of a single and a 7-day administration of ostruthin on mice that were subjected to chronic unpredictable mild stress (CUMS). A single administration of ostruthin showed antidepressive effects in the tail suspension and forced swim tests of CUMS-treated mice. Unexpectedly, the 7-day administration exhibited only insignificant antidepressive and anxiolytic effects. The 7-day regimen did not affect food intake or body-weight gain, suggesting the absence of apparent cytotoxicity. The mice receiving the 7-day administration had significantly lower blood concentrations of ostruthin compared to those receiving a single dose, suggesting an upregulation of drug-metabolizing activities. These findings suggest that there is a need for stable TREK-1 channel activators that are not affected by drug metabolism.

Discovery of pyrazolopyrimidines that selectively inhibit CSF-1R kinase by iterative design, synthesis and screening against glioblastoma cells

Glioblastoma multiforme (GBM) is the most aggressive type of brain cancer in adults, with an average life expectancy under treatment of approx. 15 months. GBM is characterised by a complex set of genetic alterations that results in significant disruption of receptor tyrosine kinase (RTK) signaling. We report here an exploration of the pyrazolo[3,4-d]pyrimidine scaffold in search for antiproliferative compounds directed to GBM treatment. Small compound libraries were synthesised and screened against GBM cells to build up structure-antiproliferative activity-relationships (SAARs) and inform further rounds of design, synthesis and screening. 76 novel compounds were generated through this iterative process that found low micromolar potencies against selected GBM lines, including patient-derived stem cells. Phenomics analysis demonstrated preferential activity against glioma cells of the mesenchymal subtype, whereas kinome screening identified colony stimulating factor-1 receptor (CSF-1R) as the lead's target, a RTK implicated in the tumourigenesis and progression of different cancers and the immunoregulation of the GBM microenvironment.

The Trends and Future Prospective of In Silico Models from the Viewpoint of ADME Evaluation in Drug Discovery

Drug discovery and development are aimed at identifying new chemical molecular entities (NCEs) with desirable pharmacokinetic profiles for high therapeutic efficacy. The plasma concentrations of NCEs are a biomarker of their efficacy and are governed by pharmacokinetic processes such as absorption, distribution, metabolism, and excretion (ADME). Poor ADME properties of NCEs are a major cause of attrition in drug development. ADME screening is used to identify and optimize lead compounds in the drug discovery process. Computational models predicting ADME properties have been developed with evolving model-building technologies from a simplified relationship between ADME endpoints and physicochemical properties to machine learning, including support vector machines, random forests, and convolution neural networks. Recently, in the field of in silico ADME research, there has been a shift toward evaluating the in vivo parameters or plasma concentrations of NCEs instead of using predictive results to guide chemical structure design. Another research hotspot is the establishment of a computational prediction platform to strengthen academic drug discovery. Bioinformatics projects have produced a series of in silico ADME models using free software and open-access databases. In this review, we introduce prediction models for various ADME parameters and discuss the currently available academic drug discovery platforms.