The rise of deep learning in drug discovery

Probabilistic Design Space Exploration and Optimization via Bayesian Approach for a Fluid Bed Drying Process

The concept of Design Space (DS), delineated as a region of investigated variables aimed at maintaining product quality, was introduced in the International Conference on Harmonisation (ICH) Q8 as a framework to direct pharmaceutical development. However, the complexity of processes and the presence of uncertainties in pharmaceutical manufacturing increase the difficulties of exploring a reliable and robust DS. This study investigates the probabilistic design space to explain the process operability and performance reliability using a Bayesian approach for a fluid bed drying process. We initially developed a Bayesian model by integrating a surrogate-based predictive regression model with an uncertainty component (attributed to material variability). Subsequently, employing a grid search-based technique to discretize the operational variable domain facilitates the exploration of the probabilistic DS to meet the specified product quality requirements. Meanwhile, optimization is employed to obtain the maximum DS region and enhance its operability. Results demonstrate that the Bayesian approach is an effective method to identify a probability DS to guarantee product quality at the desired reliability level considering material and process uncertainty.

Automatic cerebral microbleeds detection from MR images via multi-channel and multi-scale CNNs

BACKGROUND

Computer-aided detection (CAD) systems have been widely used to assist medical professionals in interpreting medical images, aiding in the detection of potential diseases. Despite their usefulness, CAD systems cannot yet fully replace doctors in diagnosing many conditions due to limitations in current algorithms. Cerebral microbleeds (CMBs) are a critical area of concern for neurological health, and accurate detection of CMBs is essential for understanding their impact on brain function. This study aims to improve CMB detection by enhancing existing machine learning algorithms.

METHODS

This paper presents four CNN-based algorithms designed to enhance CMB detection. The detection methods are categorized into traditional machine learning approaches and deep learning-based methods. The traditional methods, while computationally efficient, offer lower sensitivity, while CNN-based approaches promise greater accuracy. The algorithms proposed in this study include a multi-channel CNN with optimized architecture and a multiscale CNN structure, both of which were designed to reduce false positives and improve overall performance.

RESULTS

The first CNN algorithm, with an optimized multi-channel architecture, demonstrated a sensitivity of 99.6 %, specificity of 99.3 %, and accuracy of 99.5 %. The fourth algorithm, based on a stable multiscale CNN structure, achieved sensitivity of 98.2 %, specificity of 97.4 %, and accuracy of 97.8 %. Both algorithms exhibited a significant reduction in false positives compared to traditional methods. The experiments conducted confirm the effectiveness of these algorithms in improving the precision and reliability of CMB detection.

CONCLUSION

The proposed CNN-based algorithms demonstrate a significant advancement in the automated detection of CMBs, with notable improvements in sensitivity, specificity, and accuracy. These results underscore the potential of deep learning models, particularly CNNs, in enhancing CAD systems for neurological disease detection and reducing diagnostic errors. Further research and optimization may allow these algorithms to be integrated into clinical practices, providing more reliable support for healthcare professionals.

MHNfs: Prompting In-Context Bioactivity Predictions for Low-Data Drug Discovery

Today's drug discovery increasingly relies on computational and machine learning approaches to identify novel candidates, yet data scarcity remains a significant challenge. To address this limitation, we present MHNfs, an application specifically designed to predict molecular activity in low-data scenarios. At its core, MHNfs leverages a state-of-the-art few-shot activity prediction model, named MHNfs, which has demonstrated strong performance across a large set of prediction tasks in the benchmark data set FS-Mol. The application features an intuitive interface that enables users to prompt the model for precise activity predictions based on a small number of known active and inactive molecules, akin to interactive interfaces for large language models. To evaluate its efficacy, we simulate real-world scenarios by recasting PubChem bioassays as few-shot prediction tasks. MHNfs offers a streamlined and accessible solution for deploying advanced few-shot learning models, providing a valuable tool for accelerating drug discovery.

Graph-Aware AURALSTM: An Attentive Unified Representation Architecture with BiLSTM for Enhanced Molecular Property Prediction

Predicting molecular properties with high accuracy is essential across scientific fields, from drug discovery and biotechnology to materials science and environmental research. In biomedical sciences, accurate molecular property prediction is crucial for elucidating disease mechanisms, identifying potential drug candidates, and optimising various processes. However, existing approaches, often based on low-dimensional representations, fail to capture the intricate spatial and structural complexities of molecular data. This study introduces a novel hybrid deep learning model, the Graph-Aware AURA-LSTM (Attentive Unified Representation Architecture-Long Short-Term Memory), designed to determine molecular properties with unprecedented accuracy using advanced graphical representations. AURA-LSTM combines multiple Graph Neural Network (GNN) architectures, specifically Graph Convolutional Networks (GCNs), Graph Attention Networks (GATs), and Graph Isomorphism Networks (GINs), in a parallel structure to comprehensively capture the multidimensional structural features of molecules. Within this architecture, GCNs incorporate local structural relationships, GATs apply attention mechanisms to highlight critical structural elements, and GINs capture intricate molecular details through isomorphic distinction, resulting in a richly detailed feature matrix. The feature layer then processes this BiLSTM matrix, which evaluates temporal relationships to enhance molecular feature classification. Evaluated on eight benchmark datasets, AURA-LSTM demonstrated superior performance, consistently achieving over 90% accuracy and outperforming state-of-the-art methods. These results position AURA-LSTM as a robust tool for molecular feature classification, uniquely capable of integrating temporally aware insights from distinct GNN architectures.

Capsule neural network and its applications in drug discovery

Deep learning holds great promise in drug discovery, yet its application is hindered by high labeling costs and limited datasets. Developing algorithms that effectively learn from sparsely labeled data is crucial. Capsule networks (CapsNet), introduced in 2017, solve the spatial information loss in traditional neural networks and excel in handling small datasets by capturing spatial hierarchical relationships among features. This capability makes CapsNet particularly promising for drug discovery, where data scarcity is a common challenge. Various modified CapsNet architectures have been successfully applied to drug design and discovery tasks. This review provides a comprehensive analysis of CapsNet's theoretical foundations, its current applications in drug discovery, and its performance in addressing key challenges in the field. Additionally, the study highlights the limitations of CapsNet and outlines potential future research directions to further enhance its utility in drug discovery, offering valuable insights for researchers in both computational and pharmaceutical sciences.

Design and Synthesis of Magnolol Derivatives Using Integrated CNNs and Pharmacophore Approaches for Enhanced Parasiticidal Activity in Aquaculture

Aquaculture is a rapidly growing sector of global food production, playing a vital role in poverty alleviation, food security, and income generation. However, it faces substantial challenges, particularly due to infections caused by the protozoan Ichthyophthirius multifiliis, leading to reduced yields and economic losses. In this study, two classes of magnolol derivatives (M1-M26) were synthesized and evaluated for their parasiticidal activity against I. multifiliis theronts and tomonts. Based on the determined EC50 values, two significant convolutional neural networks (CNNs) and pharmacophore models were developed, guiding the design and synthesis of three new compounds. Among these, compound new-1 exhibited superior parasiticidal efficacy both in vitro and in vivo compared to magnolol and the commercial parasiticide methylene blue, with EC50 values of 0.197 and 0.252 mg/L, respectively. Toxicity assays demonstrated that compound new-1 showed no significant harmful effects on goldfish and zebrafish embryos at effective concentrations. Further, in silico and experimental analyses identified CDK2 as a molecular target for compound new-1, which disrupted the reproduction and exerts parasiticidal activity against I. multifiliis. Additionally, compound new-1 exhibited antifungal activity against Aspergillus flavus with a MIC of ∼0.8 mg/mL, suggesting its potential application in food preservation. This study provides valuable insights into the development of magnolol-based agents for aquaculture.

Machine Learning-Enhanced Network Pharmacology in Traditional Chinese Medicine: Mechanistic Insights Into Chai Hu Gui Zhi Tang for Allergic Rhinitis

Network pharmacology has become a widely used approach for studying complex herbal medicines. This method helps researchers identify potentially active compounds and mechanisms, which can then guide further experiments. However, current network pharmacology methods often face issues like inconsistent compound records and unreliable screening standards, leading to inaccurate results. To address these challenges, we developed an improved workflow using Chai Hu Gui Zhi Tang (CHGZT) as an example. First, we used advanced analytical techniques (ultra-high-performance liquid chromatography quadrupole time-of-flight mass spectrometry) to rapidly identify chemical components in the herbal formula. Next, we created a machine learning model to predict compounds with anti-allergic rhinitis activity, allowing systematic selection of key components for network analysis. Our results showed that specific compounds like cinnamic acid and citric acid may combat allergic rhinitis by regulating immune-related genes (interleukin [IL]-4 and IL-5) while influencing biological processes such as "stress response" and "metabolism of foreign substances." These findings confirm the effectiveness of our optimized method and highlight CHGZT's potential as a therapeutic option for allergic rhinitis.

A beginner's approach to deep learning applied to VS and MD techniques

It has become impossible to imagine the fields of biochemistry and medicinal chemistry without computational chemistry and molecular modelling techniques. In many steps of the drug development process in silico methods have become indispensable. Virtual screening (VS) can tremendously expedite the early discovery phase, whilst the use of molecular dynamics (MD) simulations forms a powerful additional tool to in vitro methods throughout the entire drug discovery process. In the field of biochemistry, MD has also become a compelling method for studying biophysical systems (e.g., protein folding) complementary to experimental techniques. However, both VS and MD come with their own limitations and methodological difficulties, from hardware limitations to restrictions in algorithmic capabilities. One solution to overcoming these difficulties lies in the field of machine learning (ML), and more specifically deep learning (DL). There are many ways in which DL can be applied to these molecular modelling techniques to achieve more accurate results in a more efficient manner or expedite the data analysis of the acquired results. Despite steadily increasing interest in DL amidst computational chemists, knowledge is still limited and scattered over different resources. This review is aimed at computational chemists with knowledge of molecular modelling, who wish to possibly integrate DL approaches in their research and already have a basic understanding of the fundamentals of DL. This review focusses on a survey of recent applications of DL in molecular modelling techniques. The different sections are logically subdivided, based on where DL is integrated in the research: (1) for the improvement of VS workflows, (2) for the improvement of certain workflows in MD simulations, (3) for aiding in the calculations of interatomic forces, or (4) for data analysis of MD trajectories. It will become clear that DL has the capacity to completely transform the way molecular modelling is carried out.

Rapid traversal of vast chemical space using machine learning-guided docking screens

The accelerating growth of make-on-demand chemical libraries provides unprecedented opportunities to identify starting points for drug discovery with virtual screening. However, these multi-billion-scale libraries are challenging to screen, even for the fastest structure-based docking methods. Here we explore a strategy that combines machine learning and molecular docking to enable rapid virtual screening of databases containing billions of compounds. In our workflow, a classification algorithm is trained to identify top-scoring compounds based on molecular docking of 1 million compounds to the target protein. The conformal prediction framework is then used to make selections from the multi-billion-scale library, reducing the number of compounds to be scored by docking. The CatBoost classifier showed an optimal balance between speed and accuracy and was used to adapt the workflow for screens of ultralarge libraries. Application to a library of 3.5 billion compounds demonstrated that our protocol can reduce the computational cost of structure-based virtual screening by more than 1,000-fold. Experimental testing of predictions identified ligands of G protein-coupled receptors and demonstrated that our approach enables discovery of compounds with multi-target activity tailored for therapeutic effect.

Artificial intelligence in anti-obesity drug discovery: unlocking next-generation therapeutics

Obesity, a multifactorial disease linked to severe health risks, requires innovative treatments beyond lifestyle changes and current medications. Existing anti-obesity drugs face limitations regarding efficacy, side effects, weight regain and high costs. Artificial intelligence (AI) is emerging as a pivotal tool in drug discovery, expediting the identification of novel drug candidates and optimizing treatment strategies. This review examines AI's potential in developing next-generation anti-obesity therapeutics, with a focus on glucagon-like peptide-1 receptor agonists (GLP-1 RAs) and their role in discovering anti-obesity peptides. Additionally, it explores integration challenges and offers future perspectives on leveraging AI to reshape the landscape of anti-obesity drug discovery.

Perspectives: Comparison of deep learning segmentation models on biophysical and biomedical data

Deep learning-based approaches are now widely used across biophysics to help automate a variety of tasks including image segmentation, feature selection, and deconvolution. However, the presence of multiple competing deep learning architectures, each with its own advantages and disadvantages, makes it challenging to select an architecture best suited for a specific application. As such, we present a comprehensive comparison of common models. Here, we focus on the task of segmentation assuming typical (often small) training dataset sizes available from biophysics experiments and compare the following four commonly used architectures: convolutional neural networks, U-Nets, vision transformers, and vision state space models. In doing so, we establish criteria for determining optimal conditions under which each model excels, thereby offering practical guidelines for researchers and practitioners in the field.

Publishing neural networks in drug discovery might compromise training data privacy

This study investigates the risks of exposing confidential chemical structures when machine learning models trained on these structures are made publicly available. We use membership inference attacks, a common method to assess privacy that is largely unexplored in the context of drug discovery, to examine neural networks for molecular property prediction in a black-box setting. Our results reveal significant privacy risks across all evaluated datasets and neural network architectures. Combining multiple attacks increases these risks. Molecules from minority classes, often the most valuable in drug discovery, are particularly vulnerable. We also found that representing molecules as graphs and using message-passing neural networks may mitigate these risks. We provide a framework to assess privacy risks of classification models and molecular representations, available at https://github.com/FabianKruger/molprivacy . Our findings highlight the need for careful consideration when sharing neural networks trained on proprietary chemical structures, informing organisations and researchers about the trade-offs between data confidentiality and model openness.

Machine Learning-Guided Screening and Molecular Docking for Proposing Naturally Derived Drug Candidates Against MERS-CoV 3CL Protease

In this study, we utilized machine learning techniques to identify potential inhibitors of the MERS-CoV 3CL protease. Among the models evaluated, the Random Forest (RF) algorithm exhibited the highest predictive performance, achieving an accuracy of 0.97, an ROC-AUC score of 0.98, and an F1-score of 0.98. Following model validation, we applied it to a dataset of 14,194 naturally occurring compounds from PubChem. The top-ranked compounds were subsequently subjected to molecular docking, which identified Perenniporide B, Phellifuropyranone A, and Terrestrol G as the most promising candidates, with binding energies of -9.17, -9.08, and -8.71 kcal/mol, respectively. These compounds formed strong interactions with key catalytic residues, suggesting significant inhibitory potential against the viral protease. Furthermore, molecular dynamics simulations confirmed their stability within the active site, reinforcing their viability as antiviral agents. This study demonstrates the effectiveness of integrating machine learning with molecular modeling to accelerate the discovery of therapeutic candidates against emerging viral threats.

Integrating artificial intelligence in drug discovery and early drug development: a transformative approach

Artificial intelligence (AI) can transform drug discovery and early drug development by addressing inefficiencies in traditional methods, which often face high costs, long timelines, and low success rates. In this review we provide an overview of how to integrate AI to the current drug discovery and development process, as it can enhance activities like target identification, drug discovery, and early clinical development. Through multiomics data analysis and network-based approaches, AI can help to identify novel oncogenic vulnerabilities and key therapeutic targets. AI models, such as AlphaFold, predict protein structures with high accuracy, aiding druggability assessments and structure-based drug design. AI also facilitates virtual screening and de novo drug design, creating optimized molecular structures for specific biological properties. In early clinical development, AI supports patient recruitment by analyzing electronic health records and improves trial design through predictive modeling, protocol optimization, and adaptive strategies. Innovations like synthetic control arms and digital twins can reduce logistical and ethical challenges by simulating outcomes using real-world or virtual patient data. Despite these advancements, limitations remain. AI models may be biased if trained on unrepresentative datasets, and reliance on historical or synthetic data can lead to overfitting or lack generalizability. Ethical and regulatory issues, such as data privacy, also challenge the implementation of AI. In conclusion, in this review we provide a comprehensive overview about how to integrate AI into current processes. These efforts, although they will demand collaboration between professionals, and robust data quality, have a transformative potential to accelerate drug development.

Clinical and Operational Applications of Artificial Intelligence and Machine Learning in Pharmacy: A Narrative Review of Real-World Applications

Over the past five years, the application of artificial intelligence (AI) including its significant subset, machine learning (ML), has significantly advanced pharmaceutical procedures in community pharmacies, hospital pharmacies, and pharmaceutical industry settings. Numerous notable healthcare institutions, such as Johns Hopkins University, Cleveland Clinic, and Mayo Clinic, have demonstrated measurable advancements in the use of artificial intelligence in healthcare delivery. Community pharmacies have seen a 40% increase in drug adherence and a 55% reduction in missed prescription refills since implementing artificial intelligence (AI) technologies. According to reports, hospital implementations have reduced prescription distribution errors by up to 75% and enhanced the detection of adverse medication reactions by up to 65%. Numerous businesses, such as Atomwise and Insilico Medicine, assert that they have made noteworthy progress in the creation of AI-based medical therapies. Emerging technologies like federated learning and quantum computing have the potential to boost the prediction of protein-drug interactions by up to 300%, despite challenges including high implementation costs and regulatory compliance. The significance of upholding patient-centred care while encouraging technology innovation is emphasised in this review.

Achieving well-informed decision-making in drug discovery: a comprehensive calibration study using neural network-based structure-activity models

In the drug discovery process, where experiments can be costly and time-consuming, computational models that predict drug-target interactions are valuable tools to accelerate the development of new therapeutic agents. Estimating the uncertainty inherent in these neural network predictions provides valuable information that facilitates optimal decision-making when risk assessment is crucial. However, such models can be poorly calibrated, which results in unreliable uncertainty estimates that do not reflect the true predictive uncertainty. In this study, we compare different metrics, including accuracy and calibration scores, used for model hyperparameter tuning to investigate which model selection strategy achieves well-calibrated models. Furthermore, we propose to use a computationally efficient Bayesian uncertainty estimation method named HMC Bayesian Last Layer (HBLL), which generates Hamiltonian Monte Carlo (HMC) trajectories to obtain samples for the parameters of a Bayesian logistic regression fitted to the hidden layer of the baseline neural network. We report that this approach improves model calibration and achieves the performance of common uncertainty quantification methods by combining the benefits of uncertainty estimation and probability calibration methods. Finally, we show that combining post hoc calibration method with well-performing uncertainty quantification approaches can boost model accuracy and calibration.

MGDM: Molecular generation using a multinomial diffusion model

Accurate analysis of molecular structures and the rapid generation of valid molecules remain significant challenges in De Novo drug design. In this study, we propose the Multinomial Generated Diffusion Model (MGDM) for molecular generation. This model leverages a multinomial diffusion framework to process discrete data, with a focus on learning the multinomial distribution inherent in the dataset. During the generation process, the model progressively denoises molecules, transitioning from a uniform noise distribution to ultimately produce valid molecular structures. Initially, we generate molecules unconditionally to expand the compound library. In the next phase, we focus on generating molecules with specific properties to assess the model's capacity for conditional generation. For this, we implement a classifier-free guidance strategy, which directs the diffusion model's task without the need for training separate classifier models. To validate the effectiveness of our framework, we conducted experiments using the Molecular Sets (MOSES) dataset. The results demonstrate that, compared to several state-of-the-art methods, MGDM generates valid molecules while achieving superior or comparable performance in terms of novelty and diversity.

Machine Learning Clustering of Water-Water Interactions in the Cambridge Structural Database

This study investigated the application of clustering machine learning techniques to analyze water-water contacts in crystal structures deposited in the Cambridge Structural Database. The initial dataset was divided into three groups regarding interaction energies between water molecules, and were separately analyzed. The application of machine learning methods enabled finding similar groups of contacts and defining their geometrical parameters. By carefully scrutinizing geometric parameters and visually examining clustering results, we demonstrated how valuable insights into the diverse spectrum of interactions between water molecules can be gained. Expanding the applicability of clustering methods can be achieved by integrating them into existing software for visualizing crystal structures. This approach has the potential to discover new types of interactions and enhance our understanding of molecular behavior.

Decoding Drug Discovery: Exploring A-to-Z In Silico Methods for Beginners

The drug development process is a critical challenge in the pharmaceutical industry due to its time-consuming nature and the need to discover new drug potentials to address various ailments. The initial step in drug development, drug target identification, often consumes considerable time. While valid, traditional methods such as in vivo and in vitro approaches are limited in their ability to analyze vast amounts of data efficiently, leading to wasteful outcomes. To expedite and streamline drug development, an increasing reliance on computer-aided drug design (CADD) approaches has merged. These sophisticated in silico methods offer a promising avenue for efficiently identifying viable drug candidates, thus providing pharmaceutical firms with significant opportunities to uncover new prospective drug targets. The main goal of this work is to review in silico methods used in the drug development process with a focus on identifying therapeutic targets linked to specific diseases at the genetic or protein level. This article thoroughly discusses A-to-Z in silico techniques, which are essential for identifying the targets of bioactive compounds and their potential therapeutic effects. This review intends to improve drug discovery processes by illuminating the state of these cutting-edge approaches, thereby maximizing the effectiveness and duration of clinical trials for novel drug target investigation.

Prediction of Drug-Target Interactions With High- Quality Negative Samples and a Network-Based Deep Learning Framework

Identification of drug-target interactions (DTIs) plays a crucial role in drug discovery. Compared to traditional experimental methods, computer-based methods for predicting DTIs can significantly reduce the time and financial burdens of drug development. In recent years, numerous machine learning-based methods have been proposed for predicting potential DTIs. However, a common limitation among these methods is the absence of high-quality negative samples. Moreover, the effective extraction of multisource information of drugs and proteins for DTI prediction remains a significant challenge. In this paper, we investigated two aspects: the selection of high-quality negative samples and the construction of a high-performance DTI prediction framework. Specifically, we found two types of hidden biases when randomly selecting negative samples from unlabeled drug-protein pairs and proposed a negative sample selection approach based on complex network theory. Furthermore, we proposed a novel DTI prediction method named HNetPa-DTI, which integrates topological information from the drug-protein-disease heterogeneous network and gene ontology (GO) and pathway annotation information of proteins. Specifically, we extracted topological information of the drug-protein-disease heterogeneous network using heterogeneous graph neural networks, and obtained GO and pathway annotation information of proteins from the GO term semantic similarity networks, GO term-protein bipartite networks, and pathway-protein bipartite network using graph neural networks. Experimental results show that HNetPa-DTI outperforms the baseline methods on four types of prediction tasks, demonstrating the superiority of our method.

An in-depth review of AI-powered advancements in cancer drug discovery

The convergence of artificial intelligence (AI) and genomics is redefining cancer drug discovery by facilitating the development of personalized and effective therapies. This review examines the transformative role of AI technologies, including deep learning and advanced data analytics, in accelerating key stages of the drug discovery process: target identification, drug design, clinical trial optimization, and drug response prediction. Cutting-edge tools such as DrugnomeAI and PandaOmics have made substantial contributions to therapeutic target identification, while AI's predictive capabilities are driving personalized treatment strategies. Additionally, advancements like AlphaFold highlight AI's capacity to address intricate challenges in drug development. However, the field faces significant challenges, including the management of large-scale genomic datasets and ethical concerns surrounding AI deployment in healthcare. This review underscores the promise of data-centric AI approaches and emphasizes the necessity of continued innovation and interdisciplinary collaboration. Together, AI and genomics are charting a path toward more precise, efficient, and transformative cancer therapeutics.

Navigating the challenges of engineering composite specialized metabolite pathways in plants

Plants are a valuable source of diverse specialized metabolites with numerous applications. However, these compounds are often produced in limited quantities, particularly under unfavorable ecological conditions. To achieve sufficient levels of target metabolites, alternative strategies such as pathway engineering in heterologous systems like microbes (e.g., bacteria and fungi) or cell-free systems can be employed. Another approach is plant engineering, which aims to either enhance the native production in the original plant or reconstruct the target pathway in a model plant system. Although increasing metabolite production in the native plant is a promising strategy, these source plants are often exotic and pose significant challenges for genetic manipulation. Effective pathway engineering requires comprehensive prior knowledge of the genes and enzymes involved, as well as the precursor, intermediate, branching, and final metabolites. Thus, a thorough elucidation of the biosynthetic pathway is closely linked to successful metabolic engineering in host or model systems. In this review, we highlight recent advances in strategies for biosynthetic pathway elucidation and metabolic engineering. We focus on efforts to engineer complex, multi-step pathways that require the expression of at least eight genes for transient and three genes for stable transformation. Reports on the engineering of complex pathways in stably transformed plants remain relatively scarce. We discuss the major hurdles in pathway elucidation and strategies for overcoming them, followed by an overview of achievements, challenges, and solutions in pathway reconstitution through metabolic engineering. Recent advances including computer-based predictions offer valuable platforms for the sustainable production of specialized metabolites in plants.

Pretraining graph transformers with atom-in-a-molecule quantum properties for improved ADMET modeling

We evaluate the impact of pretraining Graph Transformer architectures on atom-level quantum-mechanical features for the modeling of absorption, distribution, metabolism, excretion, and toxicity (ADMET) properties of drug-like compounds. We compare this pretraining strategy with two others: one based on molecular quantum properties (specifically the HOMO-LUMO gap) and one using a self-supervised atom masking technique. After fine-tuning on Therapeutic Data Commons ADMET datasets, we evaluate the performance improvement in the different models observing that models pretrained with atomic quantum mechanical properties produce in general better results. We then analyze the latent representations and observe that the supervised strategies preserve the pretraining information after fine-tuning and that different pretrainings produce different trends in latent expressivity across layers. Furthermore, we find that models pretrained on atomic quantum mechanical properties capture more low-frequency Laplacian eigenmodes of the input graph via the attention weights and produce better representations of atomic environments within the molecule. Application of the analysis to a much larger non-public dataset for microsomal clearance illustrates generalizability of the studied indicators. In this case the performances of the models are in accordance with the representation analysis and highlight, especially for the case of masking pretraining and atom-level quantum property pretraining, how model types with similar performance on public benchmarks can have different performances on large scale pharmaceutical data.Scientific contributionWe systematically compared three different data type/methodologies for pretraining molecular Graphormer with the purpose of modeling ADMET properties as downstream tasks. The learned representations from differently pretrained models were analyzed in addition to comparison of downstream task performances that have been typically reported in similar works. Such examination methodologies, including a newly introduced analysis of Graphormer's Attention Rollout Matrix, can guide pretraining strategy selection, as corroborated by a performance evaluation on a larger internal dataset.

Exploring a new paradigm for serum-accessible component rules of natural medicines using machine learning and development and validation of a direct predictive model

In the field of pharmaceutical research, Lipinski's Rule of Five (RO5) was once widely regarded as the prevailing standard for the development of novel drugs. Despite the fact that an increasing number of recently approved drugs no longer adhere to this rule, it continues to serve as a valuable guiding principle in the field of drug discovery. The present study aims to establish a set of rules specifically for the serum-accessible components of natural medicines. A comprehensive literature review was conducted to collect data on serum-accessible components of natural medicines, and machine learning methods were then applied to analyse and screen molecular features distinguishing serum-accessible components from non-serum-accessible ones. The most critical rules for serum-accessible components of natural medicines were identified, and these were named the "Natural Medicine's Rule of 5 (NMRO5)." We then compared the molecular property distributions and predictive performance of NMRO5 with RO5. Then, we developed a predictive model capable of directly assessing the possibility of a molecule being serum-accessible. This model was validated using in vivo experiments on multiple natural medicines. Furthermore, we performed molecular modifications on serum-accessible components to "violate" NMRO5, conducting both forward and reverse validations to confirm the reliability of NMRO5. The results obtained revealed that NMRO5 is characterised by the following: higher TPSA, MaxEState, and PEOE VSA1 values, and lower LogP and MinEState values. This indicates that natural medicine components with these properties are more likely to be serum-accessible or remain in plasma rather than being rapidly eliminated. The investigation revealed significant disparities among the five molecular properties of NMRO5, and the predictive performance of eight models based on NMRO5 consistently outperformed those based on RO5. This finding suggests that NMRO5 provides a more reliable framework for determining whether a molecule is serum-accessible compared to RO5. Finally, we developed a direct predictive model for serum-accessible components, achieving an accuracy of 0.7257, an F1 score of 0.7223, and an AUC of 0.7553.

Challenges and limitations of computer-aided drug design

Molecular Modelling in Drug Designing or Computer Aided Drug Designing (CADD) plays a significant role in new drug identification in the current world. However, it has sensitivity challenges and limitation because theoretical models involve assumption and approximations Computational models are not very accurate, some of the major challenges that face these models include the following. These include, for instance, molecular-docking or molecular-dynamics-simulation models which may not represent an accurate biological system and thus the predictions will be wrong. CADD depends on the availability of accurate, high-quality structural information for target proteins and ligand. Unfortunately, there are instances when experimental structures are not available, and homology models are employed, which can be imprecise. The computational cost is another drawback; only high accuracy simulations call for huge amounts of computational power and time well-suited for screening a multitude of agents. Moreover, they have weaknesses in determining pharmacokinetic and toxicity patterns of compounds that influence drug performance and effectiveness. In other words, even though CADD greatly helps drug discovery, it is still constrained by experimental validation to solve its drawbacks and optimize its foretelling.

Advanced AI and ML frameworks for transforming drug discovery and optimization: With innovative insights in polypharmacology, drug repurposing, combination therapy and nanomedicine

Artificial Intelligence (AI) and Machine Learning (ML) are transforming drug discovery by overcoming traditional challenges like high costs, time-consuming, and frequent failures. AI-driven approaches streamline key phases, including target identification, lead optimization, de novo drug design, and drug repurposing. Frameworks such as deep neural networks (DNNs), convolutional neural networks (CNNs), and deep reinforcement learning (DRL) models have shown promise in identifying drug targets, optimizing delivery systems, and accelerating drug repurposing. Generative adversarial networks (GANs) and variational autoencoders (VAEs) aid de novo drug design by creating novel drug-like compounds with desired properties. Case studies, such as DDR1 kinase inhibitors designed using generative models and CDK20 inhibitors developed via structure-based methods, highlight AI's ability to produce highly specific therapeutics. Models like SNF-CVAE and DeepDR further advance drug repurposing by uncovering new therapeutic applications for existing drugs. Advanced ML algorithms enhance precision in predicting drug efficacy, toxicity, and ADME-Tox properties, reducing development costs and improving drug-target interactions. AI also supports polypharmacology by optimizing multi-target drug interactions and enhances combination therapy through predictions of drug synergies and antagonisms. In nanomedicine, AI models like CURATE.AI and the Hartung algorithm optimize personalized treatments by predicting toxicological risks and real-time dosing adjustments with high accuracy. Despite its potential, challenges like data quality, model interpretability, and ethical concerns must be addressed. High-quality datasets, transparent models, and unbiased algorithms are essential for reliable AI applications. As AI continues to evolve, it is poised to revolutionize drug discovery and personalized medicine, advancing therapeutic development and patient care.

Enhancing lipid identification in LC-HRMS data through machine learning-based retention time prediction

The comprehensive identification of peaks in untargeted lipidomics using LC-MS/MS remains a significant challenge. Confidence in lipid annotation can be greatly improved by integrating a highly accurate machine learning-based retention time prediction model. Such an approach enables the identification of lipids for understanding pathogenic mechanisms, biomarker discovery, and drug screening. In this study, we developed a machine learning model to predict retention times and facilitate lipid peak annotations in LC-MS-based untargeted lipidomics. Our model achieved high correlation coefficients of 0.998 and 0.990, with mean absolute errors (MAE) of 0.107 min and 0.240 min for the training and test sets, respectively. External validation showed similarly strong performance, with correlations of 0.991 and 0.978, and MAE values of 0.241 min and 0.270 min. We also compared the impact of molecular descriptors and molecular fingerprints on the model's performance, finding that molecular descriptors outperformed molecular fingerprints across all datasets when using Random Forest (RF) for model construction. Notably, this retention time calibration model demonstrates robust performance across chromatographic systems with comparable gradients and flow rates. Overall, this machine learning model enhances lipid annotation accuracy and reduces errors in untargeted lipidomics, improving data analysis across multiple datasets.

Deep learning: A game changer in drug design and development

The lengthy and costly drug discovery process is transformed by deep learning, a subfield of artificial intelligence. Deep learning technologies expedite the procedure, increasing treatment success rates and speeding life-saving procedures. Deep learning stands out in target identification and lead selection. Deep learning greatly accelerates initial stage by analyzing large datasets of biological data to identify possible therapeutic targets and rank targeted drug molecules with desired features. Predicting possible adverse effects is another significant challenge. Deep learning offers prompt and efficient assistance with toxicology prediction in a very short time, deep learning algorithms can forecast a new drug's possible harm. This enables to concentrate on safer alternatives and steer clear of late-stage failures brought on by unanticipated toxicity. Deep learning unlocks the possibility of drug repurposing; by examining currently available medications, it is possible to find whole new therapeutic uses. This method speeds up development of diseases that were previously incurable. De novo drug discovery is made possible by deep learning when combined with sophisticated computational modeling, it can create completely new medications from the ground. Deep learning can recommend and direct towards new drug candidates with high binding affinities and intended therapeutic effects by examining molecular structures of disease targets. This provides focused and personalized medication. Lastly, drug characteristics can be optimized with aid of deep learning. Researchers can create medications with higher bioavailability and fewer toxicity by forecasting drug pharmacokinetics. In conclusion, deep learning promises to accelerate drug development, reduce costs, and ultimately save lives.

Machine Learning-Guided Prediction of Hydroformylation

A holistic model for predicting yield and linear selectivity for the hydroformylation of 1-octene was developed by machine learning using the experimental data collected from literatures. Physical organic chemistry (POC) parameter-based descriptors were adopted to represent pre-catalyst molecular features. Machine learning models trained respectively by Random Forests (RF) and Extreme Gradient Boost (XGBoost) algorithm showed remarkable performance on predicting linear selectivity. The method can also comprehensively map the correlation between reaction conditions and the results. The accuracy of the prediction results was verified by experimental data.

Drug repositioning for Parkinson's disease: An emphasis on artificial intelligence approaches

Parkinson's disease (PD) is one of the most incapacitating neurodegenerative diseases (NDDs). PD is the second most common NDD worldwide which affects approximately 1-2 percent of people over 65 years. It is an attractive pursuit for artificial intelligence (AI) to contribute to and evolve PD treatments through drug repositioning by repurposing existing drugs, shelved drugs, or even candidates that do not meet the criteria for clinical trials. A search was conducted in three databases Web of Science, Scopus, and PubMed. We reviewed the data related to the last years (1975-present) to identify those drugs currently being proposed for repositioning in PD. Moreover, we reviewed the present status of the computational approach, including AI/Machine Learning (AI/ML)-powered pharmaceutical discovery efforts and their implementation in PD treatment. It was found that the number of drug repositioning studies for PD has increased recently. Repositioning of drugs in PD is taking off, and scientific communities are increasingly interested in communicating its results and finding effective treatment alternatives for PD. A better chance of success in PD drug discovery has been made possible due to AI/ML algorithm advancements. In addition to the experimentation stage of drug discovery, it is also important to leverage AI in the planning stage of clinical trials to make them more effective. New AI-based models or solutions that increase the success rate of drug development are greatly needed.

Mfgnn: Multi-Scale Feature-Attentive Graph Neural Networks for Molecular Property Prediction

In the realm of artificial intelligence-driven drug discovery (AIDD), accurately predicting the influence of molecular structures on their properties is a critical research focus. While deep learning models based on graph neural networks (GNNs) have made significant advancements in this area, prior studies have primarily concentrated on molecule-level representations, often neglecting the impact of functional group structures and the potential relationships between fragments on molecular property predictions. To address this gap, we introduce the multi-scale feature attention graph neural network (MfGNN), which enhances traditional atom-based molecular graph representations by incorporating fragment-level representations derived from chemically synthesizable BRICS fragments. MfGNN not only effectively captures both the structural information of molecules and the features of functional groups but also pays special attention to the potential relationships between fragments, exploring how they collectively influence molecular properties. This model integrates two core mechanisms: a graph attention mechanism that captures embeddings of molecules and functional groups, and a feature extraction module that systematically processes BRICS fragment-level features to uncover relationships among the fragments. Our comprehensive experiments demonstrate that MfGNN outperforms leading machine learning and deep learning models, achieving state-of-the-art performance in 8 out of 11 learning tasks across various domains, including physical chemistry, biophysics, physiology, and toxicology. Furthermore, ablation studies reveal that the integration of multi-scale feature information and the feature extraction module enhances the richness of molecular features, thereby improving the model's predictive capabilities.

Development and experimental validation of a machine learning model for the prediction of new antimalarials

A large set of antimalarial molecules (N ~ 15k) was employed from ChEMBL to build a robust random forest (RF) model for the prediction of antiplasmodial activity. Rather than depending on high throughput screening (HTS) data, molecules tested at multiple doses against blood stages of Plasmodium falciparum were used for model development. The open-access and code-free KNIME platform was used to develop a workflow to train the model on 80% of data (N ~ 12k). The hyperparameter values were optimized to achieve the highest predictive accuracy with nine different molecular fingerprints (MFPs), among which Avalon MFPs (referred to as RF-1) provided the best results. RF-1 displayed 91.7% accuracy, 93.5% precision, 88.4% sensitivity and 97.3% area under the Receiver operating characteristic (AUROC) for the remaining 20% test set. The predictive performance of RF-1 was comparable to that of the malaria inhibitor prediction platform (MAIP), a recently reported consensus model based on a large proprietary dataset. However, hits obtained from RF-1 and MAIP from a commercial library did not overlap, suggesting that these two models are complementary. Finally, RF-1 was used to screen small molecules under clinical investigations for repurposing. Six molecules were purchased, out of which two human kinase inhibitors were identified to have single-digit micromolar antiplasmodial activity. One of the hits (compound 1) was a potent inhibitor of β-hematin, suggesting the involvement of parasite hemozoin (Hz) synthesis in the parasiticidal effect. The training and test sets are provided as supplementary information, allowing others to reproduce this work.

A bibliometric analysis of studies on artificial intelligence in neuroscience

The incorporation of artificial intelligence (AI) into neuroscience has the potential to significantly enhance our comprehension of brain function and facilitate more effective diagnosis and treatment of neurological disorders. Artificial intelligence (AI) techniques, particularly deep learning and machine learning, offer transformative solutions by improving the analysis of complex neural data, facilitating early diagnosis, and enabling personalized treatment approaches. A bibliometric analysis is a method that employs quantitative techniques for the examination of scientific literature, with the objective of identifying trends in research, evaluating the impact of influential studies, and mapping the networks of collaboration. In light of the accelerated growth and interdisciplinary scope of AI applications in neuroscience, a bibliometric analysis is vital for mapping the landscape, identifying pivotal contributions, and underscoring emerging areas of interest. This study aims to address this need by examining 1,208 studies published between 1983 and 2024 from the Web of Science database. The analysis reveals a notable surge in publications since the mid-2010s, with substantial advancements in neurological imaging, brain-computer interfaces (BCI), and the diagnosis and treatment of neurological diseases. The analysis underscores the pioneering role of countries such as the United States, China, and the United Kingdom in this field and highlights the prevalence of international collaboration. This study offers a comprehensive overview of the current state and future directions of AI applications in neuroscience, as well as an examination of the transformative potential of AI in advancing neurological research and healthcare. It is recommended that future research address the ethical issues, data privacy concerns, and interpretability of AI models in order to fully capitalize on the benefits of AI in neuroscience.

Deep learning-based discovery of compounds for blood pressure lowering effects

The hypotensive side effects caused by drugs during their use have been a vexing issue. Recent studies have found that deep learning can effectively predict the biological activity of compounds by mining patterns and rules in the data, providing a potential solution for identifying drug side effects. In this study, we established a deep learning-based predictive model, utilizing a data set comprised of compounds known to either elevate or lower blood pressure. Subsequently, the trained model was used to predict the blood pressure-lowering effects of 26,000 compounds. Based on the predicted results, we randomly selected 50 molecules for validation and compared them with literature reports. The results showed that the predictions for 30 molecules were consistent with literature reports, with known antihypertensive drugs such as reserpine, guanethidine, and mecamylamine ranking at the top. We further selected 10 of these molecules and 3 related protein targets for molecular docking, and the docking results indirectly confirmed the model's accuracy. Ultimately, we discovered and validated that salaprinol significantly inhibits ACE1 activity and lowers canine blood pressure. In summary, we have established a highly accurate activity prediction model and confirmed its accuracy in predicting potential blood pressure-lowering compounds, which is expected to help patients avoid hypotensive side effects during clinical medication and also provide significant assistance in the discovery of antihypertensive drugs.

Drug Repurposing: Insights into Current Advances and Future Applications

Drug development is a complex and expensive process that involves extensive research and testing before a new drug can be approved for use. This has led to a limited availability of potential therapeutics for many diseases. Despite significant advances in biomedical science, the process of drug development remains a bottleneck, as all hypotheses must be tested through experiments and observations, which can be timeconsuming and costly. To address this challenge, drug repurposing has emerged as an innovative strategy for finding new uses for existing medications that go beyond their original intended use. This approach has the potential to speed up the drug development process and reduce costs, making it an attractive option for pharmaceutical companies and researchers alike. It involves the identification of existing drugs or compounds that have the potential to be used for the treatment of a different disease or condition. This can be done through a variety of approaches, including screening existing drugs against new disease targets, investigating the biological mechanisms of existing drugs, and analyzing data from clinical trials and electronic health records. Additionally, repurposing drugs can lead to the identification of new therapeutic targets and mechanisms of action, which can enhance our understanding of disease biology and lead to the development of more effective treatments. Overall, drug repurposing is an exciting and promising area of research that has the potential to revolutionize the drug development process and improve the lives of millions of people around the world. The present review provides insights on types of interaction, approaches, availability of databases, applications and limitations of drug repurposing.

Current Status and Future of Artificial Intelligence in Medicine

Artificial intelligence (AI) has rapidly emerged as a transformative force in medicine, revolutionizing various aspects of healthcare from diagnostics and treatment to public health and patient care. This narrative review synthesizes evidence from diverse study designs, exploring the current and future applications of AI in medicine. We highlight AI's role in improving diagnostic accuracy, optimizing treatment strategies, and enhancing patient care through personalized interventions and remote monitoring, drawing upon recent advancements and landmark studies. Emerging trends such as explainable AI and federated learning are also examined. While acknowledging the tremendous potential of AI in medicine, the review also addresses the barriers and ethical challenges that need to be overcome, including concerns about algorithmic bias, transparency, over-reliance, and the potential impact on the healthcare workforce. We emphasize the importance of establishing regulatory guidelines, fostering collaboration between clinicians and AI developers, and ensuring ongoing education for healthcare professionals. Despite these challenges, the future of AI in medicine holds immense promise, with the potential to significantly improve patient outcomes, transform healthcare delivery, and address healthcare disparities.

Physics-informed neural networks for biopharmaceutical cultivation processes: Consideration of varying process parameter settings

We present a new modeling approach for the study and prediction of important process outcomes of biotechnological cultivation processes under the influence of process parameter variations. Our model is based on physics-informed neural networks (PINNs) in combination with kinetic growth equations. Using Taylor series, multivariate external process parameter variations for important variables such as temperature, seeding cell density and feeding rates can be integrated into the corresponding kinetic rates and the governing growth equations. In addition to previous approaches, PINNs also allow continuous and differentiable functions as predictions for the process outcomes. Accordingly, our results show that PINNs in combination with Taylor-series expansions for kinetic growth equations provide a very high prediction accuracy for important process variables such as cell densities and concentrations as well as a detailed study of individual and combined parameter influences. Furthermore, the proposed approach can also be used to evaluate the outcomes of new parameter variations and combinations, which enables a saving of experiments in combination with a model-driven optimization study of the design space.

Numerical modeling and simulation for microneedles drug delivery: A novel comprehensive swelling-obstruction-mechanics model

Hydrogel microneedles have attracted significant attention in drug delivery due to their non-invasiveness and efficient administration. However, a thorough understanding of the drug transport mechanism is essential to achieve controlled drug delivery and geometry optimization of microneedles. In this study, a new swelling-obstruction-mechanics model is presented to describe the swelling and drug release behavior of hydrogel microneedles. The model integrates the swelling kinetics, the obstruction scaling of drug molecules, and the mechanical properties of hydrogel and skin and reveals the effects of swelling of the microneedle matrix and drug molecules on drug release. Subsequently, numerical simulations were conducted using the model, which enabled the optimization of hydrogel microneedle design parameters by adjusting the input variables. The results show that the geometric parameters of microneedles, especially the cross-sectional shape, have a significant effect on the drug release performance. Nevertheless, the parameters affect each other and need to be considered in the selection of a variety of factors. Additionally, penetration depth significantly affects drug release efficiency, highlighting the need for auxiliary application devices. In summary, the model advances both theoretical understanding and practical design of hydrogel microneedles, identifying key factors in drug release and optimizing their efficiency and reliability for clinical applications.

PDE4D inhibitors: Opening a new era of PET diagnostics for Alzheimer's disease

As the incidence of Alzheimer's disease (AD) continues to rise, the need for an effective PET radiotracer to facilitate early diagnosis has become more pressing than ever before in modern medicine. Phosphodiesterase (PDE) is closely related to cognitive impairment and neuroinflammatory processes in AD. Current research progress shows that specific PDE4D inhibitors radioligands can bind specifically to the PDE4D enzyme in the brain, thereby showing pathology-related signal enhancement in AD animal models, indicating the potential of these ligands as effective radiotracers. At the same time, we need to pay attention to the important role computer aided drug design (CADD) plays in advancing AD drug design and PET imaging. Future research will verify the potential of these ligands in clinical applications through computer simulation techniques, providing patients with timely intervention and treatment, which is of great significance.

Discovery and Characterization of Two Selective Inhibitors for a Mu-Class Glutathione S-Transferase of 25 kDa from Taenia solium Using Computational and Bioinformatics Tools

Glutathione S-transferases (GSTs) are promising pharmacological targets for developing antiparasitic agents against helminths, as they play a key role in detoxifying cytotoxic xenobiotics and managing oxidative stress. Inhibiting GST activity can compromise parasite viability. This study reports the successful identification of two selective inhibitors for the mu-class glutathione S-transferase of 25 kDa (Ts25GST) from Taenia solium, named i11 and i15, using a computationally guided approach. The workflow involved modeling and refining the 3D structure from the sequence using the AlphaFold algorithm and all-atom molecular dynamics simulations with an explicit solvent. Representative structures from these simulations and a putative binding site with low conservation relative to human GSTs, identified via the SILCS methodology, were employed for virtual screening through ensemble docking against a commercial compound library. The two compounds were found to reduce the enzyme's activity by 50-70% under assay conditions, while showing a reduction of only 30-35% for human mu-class GSTM1, demonstrating selectivity for Ts25GST. Notable, i11 displayed competitive inhibition with CDNB, while i15 exhibited a non-competitive inhibition type.

Challenges issues and future recommendations of deep learning techniques for SARS-CoV-2 detection utilising X-ray and CT images: a comprehensive review

The global spread of SARS-CoV-2 has prompted a crucial need for accurate medical diagnosis, particularly in the respiratory system. Current diagnostic methods heavily rely on imaging techniques like CT scans and X-rays, but identifying SARS-CoV-2 in these images proves to be challenging and time-consuming. In this context, artificial intelligence (AI) models, specifically deep learning (DL) networks, emerge as a promising solution in medical image analysis. This article provides a meticulous and comprehensive review of imaging-based SARS-CoV-2 diagnosis using deep learning techniques up to May 2024. This article starts with an overview of imaging-based SARS-CoV-2 diagnosis, covering the basic steps of deep learning-based SARS-CoV-2 diagnosis, SARS-CoV-2 data sources, data pre-processing methods, the taxonomy of deep learning techniques, findings, research gaps and performance evaluation. We also focus on addressing current privacy issues, limitations, and challenges in the realm of SARS-CoV-2 diagnosis. According to the taxonomy, each deep learning model is discussed, encompassing its core functionality and a critical assessment of its suitability for imaging-based SARS-CoV-2 detection. A comparative analysis is included by summarizing all relevant studies to provide an overall visualization. Considering the challenges of identifying the best deep-learning model for imaging-based SARS-CoV-2 detection, the article conducts an experiment with twelve contemporary deep-learning techniques. The experimental result shows that the MobileNetV3 model outperforms other deep learning models with an accuracy of 98.11%. Finally, the article elaborates on the current challenges in deep learning-based SARS-CoV-2 diagnosis and explores potential future directions and methodological recommendations for research and advancement.

Precision Adverse Drug Reactions Prediction with Heterogeneous Graph Neural Network

Accurate prediction of Adverse Drug Reactions (ADRs) at the patient level is essential for ensuring patient safety and optimizing healthcare outcomes. Traditional machine learning-based methods primarily focus on predicting potential ADRs for drugs, but they often fall short of capturing the complexity of individual demographics and the variations in ADRs experienced by different people. In this study, a novel framework called Precise Adverse Drug Reaction (PreciseADR) for patient-level ADR prediction is proposed. The approach effectively integrates relations between patients and ADRs, and harnesses the power of heterogeneous Graph Neural Networks (GNNs) to address the limitations of traditional methods. Specifically, a heterogeneous graph representation of patients is constructed, encompassing nodes that represent patients, diseases, drugs, and ADRs. By leveraging edges in the graph, crucial connections are captured such as a patient being affected by diseases, taking specific drugs, and experiencing ADRs. Next, a GNN-based model is utilized to learn latent representations of the patient nodes and facilitate the propagation of information throughout the graph structure. By employing patient embeddings that consider their diseases and drugs, potential ADRs can be accurately predicted. The PreciseADR is dedicated to effectively capturing both local and global dependencies within the heterogeneous graph, allowing for the identification of subtle patterns and interactions that play a significant role in ADRs. To evaluate the performance of the approach, extensive experiments are conducted on a large-scale real-world healthcare dataset with adverse reports from the FDA Adverse Event Reporting System (FAERS). Experimental results demonstrate that the PreciseADR achieves superior predictive performance in identifying patient-level ADRs, surpassing the strongest baseline by 3.2% in AUC score and by 4.9% in Hit@10.

Overview of Computational Toxicology Methods Applied in Drug and Green Chemical Discovery

In the field of computational chemistry, computer models are quickly and cheaply constructed to predict toxicology hazards and results, with no need for test material or animals as these computational predictions are often based on physicochemical properties of chemical structures. Multiple methodologies are employed to support in silico assessments based on machine learning (ML) and deep learning (DL). This review introduces the development of computational toxicology, focusing on ML and DL and emphasizing their importance in the field of toxicology. A fine balance between target potency, selectivity, absorption, distribution, metabolism, excretion, toxicity (ADMET) and clinical safety properties should be achieved to discover a potential new drug. It is advantageous to perform virtual predictions as early as possible in drug development processes, even before a molecule is synthesized. Currently, there are numerous commercially available and free web-based programs for toxicity prediction, which can be used to construct various predictive models. The key features of the QSAR method are also outlined, and the selection of appropriate physicochemical descriptors is a prerequisite for robust predictions. In addition, examples of open-source tools applied to toxicity prediction are included, as well as examples of the application of different computational methods for the prediction of toxicity in drug design and environmental toxicology.

Autoencoder-based inverse design and surrogate-based optimization of an integrated wet granulation manufacturing process

In pharmaceutical manufacturing, integrating model-based design and optimization can be beneficial for accelerating process development. This study explores the utilization of Machine Learning (ML) techniques as a surrogate model for the optimization of a three-unit wet-granulation based flowsheet model for solid dosage form manufacturing. First, a reduced representation of a wet granulation flowsheet model is developed, incorporating a granulation and milling process, along with a novel dissolution model that accounts for the effect of particle size, porosity, and microstructure on dissolution rate. Two optimization approaches are compared, including an autoencoder-based inverse design and a surrogate-based forward optimization. Both methods address the bi-objective problem of maximizing dissolution time and product yield by identifying the optimal granulation and mill process parameters. For this case study, both approaches were effective and incurred a similar computational cost, averaging under 4 s. However, the autoencoder approach offers an advantage through dimensionality reduction, a feature not available in surrogate-based optimization. Dimensional reduction is particularly beneficial for complex process designs with numerous inputs and outputs. The lower dimensional representation helps improve process understanding through enhanced visualization of the process design space and facilitates feasibility studies involving multiple constraints. The autoencoder-based inverse design introduced in this work showcases an implementation of AI and ML in pharmaceutical process development, demonstrating the potential to enhance process efficiency and product quality in complex manufacturing scenarios.

Multi-modal networks for real-time monitoring of intracranial acoustic field during transcranial focused ultrasound therapy

BACKGROUND AND OBJECTIVE

Transcranial focused ultrasound (tFUS) is an emerging non-invasive therapeutic technology that offers new brain stimulation modality. Precise localization of the acoustic focus to the desired brain target throughout the procedure is needed to ensure the safety and effectiveness of the treatment, but acoustic distortion caused by the skull poses a challenge. Although computational methods can provide the estimated location and shape of the focus, the computation has not reached sufficient speed for real-time inference, which is demanded in real-world clinical situations. Leveraging the advantages of deep learning, we propose multi-modal networks capable of generating intracranial pressure map in real-time.

METHODS

The dataset consisted of free-field pressure maps, intracranial pressure maps, medical images, and transducer placements was obtained from 11 human subjects. The free-field and intracranial pressure maps were computed using the k-space method. We developed network models based on convolutional neural networks and the Swin Transformer, featuring a multi-modal encoder and a decoder.

RESULTS

Evaluations on foreseen data achieved high focal volume conformity of approximately 93% for both computed tomography (CT) and magnetic resonance (MR) data. For unforeseen data, the networks achieved the focal volume conformity of 88% for CT and 82% for MR. The inference time of the proposed networks was under 0.02 s, indicating the feasibility for real-time simulation.

CONCLUSIONS

The results indicate that our networks can effectively and precisely perform real-time simulation of the intracranial pressure map during tFUS applications. Our work will enhance the safety and accuracy of treatments, representing significant progress for low-intensity focused ultrasound (LIFU) therapies.

Addressing docking pose selection with structure-based deep learning: Recent advances, challenges and opportunities

Molecular docking is a widely used technique in drug discovery to predict the binding mode of a given ligand to its target. However, the identification of the near-native binding pose in docking experiments still represents a challenging task as the scoring functions currently employed by docking programs are parametrized to predict the binding affinity, and, therefore, they often fail to correctly identify the ligand native binding conformation. Selecting the correct binding mode is crucial to obtaining meaningful results and to conveniently optimizing new hit compounds. Deep learning (DL) algorithms have been an area of a growing interest in this sense for their capability to extract the relevant information directly from the protein-ligand structure. Our review aims to present the recent advances regarding the development of DL-based pose selection approaches, discussing limitations and possible future directions. Moreover, a comparison between the performances of some classical scoring functions and DL-based methods concerning their ability to select the correct binding mode is reported. In this regard, two novel DL-based pose selectors developed by us are presented.

Suitability of large language models for extraction of high-quality chemical reaction dataset from patent literature

With the advent of artificial intelligence (AI), it is now possible to design diverse and novel molecules from previously unexplored chemical space. However, a challenge for chemists is the synthesis of such molecules. Recently, there have been attempts to develop AI models for retrosynthesis prediction, which rely on the availability of a high-quality training dataset. In this work, we explore the suitability of large language models (LLMs) for extraction of high-quality chemical reaction data from patent documents. A comparative study on the same set of patents from an earlier study showed that the proposed automated approach can enhance the current datasets by addition of 26% new reactions. Several challenges were identified during reaction mining, and for some of them alternative solutions were proposed. A detailed analysis was also performed wherein several wrong entries were identified in the previously curated dataset. Reactions extracted using the proposed pipeline over a larger patent dataset can improve the accuracy and efficiency of synthesis prediction models in future.Scientific contributionIn this work we evaluated the suitability of large language models for mining a high-quality chemical reaction dataset from patent literature. We showed that the proposed approach can significantly improve the quantity of the reaction database by identifying more chemical reactions and improve the quality of the reaction database by correcting previous errors/false positives.

Mixture-of-Experts Based Dissociation Kinetic Model for De Novo Design of HSP90 Inhibitors with Prolonged Residence Time

The dissociation rate constant (koff) significantly impacts the drug potency and dosing frequency. This work proposes a powerful optimization-based framework for de novo drug design guided by koff. First, a comprehensive database containing 2,773 unique koff values is created. Based on the database, a novel generic dissociation kinetic model is developed with a mixture-of-experts architecture, enabling high-throughput predictions of koff with high accuracy. The developed model is then integrated with an optimization-based mathematical programming approach to design drug candidates with low koff. Finally, the τ-RAMD method is utilized to rigorously verify the designed potential drug candidates. In a case study, the framework successfully identified numerous new potential HSP90 inhibitor candidates, achieving a maximum 45.7% improvement in residence time (τ = 1/koff) compared to that of a known exceptional HSP90 inhibitor. These findings demonstrate the feasibility and effectiveness of the kinetics-guided optimization-based de novo drug design framework in designing drug candidates with prolonged τ.

NP-TCMtarget: a network pharmacology platform for exploring mechanisms of action of traditional Chinese medicine

The biological targets of traditional Chinese medicine (TCM) are the core effectors mediating the interaction between TCM and the human body. Identification of TCM targets is essential to elucidate the chemical basis and mechanisms of TCM for treating diseases. Given the chemical complexity of TCM, both in silico high-throughput compound-target interaction predicting models and biological profile-based methods have been commonly applied for identifying TCM targets based on the structural information of TCM chemical components and biological information, respectively. However, the existing methods lack the integration of TCM chemical and biological information, resulting in difficulty in the systematic discovery of TCM action pathways. To solve this problem, we propose a novel target identification model NP-TCMtarget to explore the TCM target path by combining the overall chemical and biological profiles. First, NP-TCMtarget infers TCM effect targets by calculating associations between herb/disease inducible gene expression profiles and specific gene signatures for 8233 targets. Then, NP-TCMtarget utilizes a constructed binary classification model to predict binding targets of herbal ingredients. Finally, we can distinguish TCM direct and indirect targets by comparing the effect targets and binding targets to establish the action pathways of herbal component-direct target-indirect target by mapping TCM targets in the biological molecular network. We apply NP-TCMtarget to the formula XiaoKeAn to demonstrate the power of revealing the action pathways of herbal formula. We expect that this novel model could provide a systematic framework for exploring the molecular mechanisms of TCM at the target level. NP-TCMtarget is available at http://www.bcxnfz.top/NP-TCMtarget.

Drug-target interaction prediction by integrating heterogeneous information with mutual attention network

BACKGROUND

Identification of drug-target interactions is an indispensable part of drug discovery. While conventional shallow machine learning and recent deep learning methods based on chemogenomic properties of drugs and target proteins have pushed this prediction performance improvement to a new level, these methods are still difficult to adapt to novel structures. Alternatively, large-scale biological and pharmacological data provide new ways to accelerate drug-target interaction prediction.

METHODS

Here, we propose DrugMAN, a deep learning model for predicting drug-target interaction by integrating multiplex heterogeneous functional networks with a mutual attention network (MAN). DrugMAN uses a graph attention network-based integration algorithm to learn network-specific low-dimensional features for drugs and target proteins by integrating four drug networks and seven gene/protein networks collected by a certain screening conditions, respectively. DrugMAN then captures interaction information between drug and target representations by a mutual attention network to improve drug-target prediction.

RESULTS

DrugMAN achieved the best performance compared with cheminformation-based methods SVM, RF, DeepPurpose and network-based deep learing methods DTINet and NeoDT in four different scenarios, especially in real-world scenarios. Compared with SVM, RF, deepurpose, DTINet, and NeoDT, DrugMAN showed the smallest decrease in AUROC, AUPRC, and F1-Score from warm-start to Both-cold scenarios. This result is attributed to DrugMAN's learning from heterogeneous data and indicates that DrugMAN has a good generalization ability. Taking together, DrugMAN spotlights heterogeneous information to mine drug-target interactions and can be a powerful tool for drug discovery and drug repurposing.