Efficient Drug Development Research Articles

Integrating AI and Deep Learning for Efficient Drug Discovery and Target Identification

Artificial intelligence (AI) shows great potential in medical diagnosis and drug analysis. Through deep learning technology, AI can automatically extract features from large amounts of complex medical data, significantly improving diagnostic accuracy and drug development efficiency. Especially in drug target discovery and antiviral peptide classification, AI technology can accelerate data processing and prediction, helping researchers identify potential therapeutic molecules more quickly and optimize the drug development process. This study proposes and validates a Deep learn-based model, deep-Avpiden, to improve the classification and discovery efficiency of antiviral peptides (AVPs). By using sequential convolutional networks (TCNs), the model outperforms traditional recurrent neural networks (RNNs) and long short-term memory networks (LSTMs) in capturing long-term dependencies and parallel computing capabilities. In terms of dataset, we used AVPs and non-AVPS samples from multiple databases, totaling 5,414 cleaned and de-weighted peptide sequences for model training after data preprocessing and embedding. The Deep-AVPiden model has been shown to outperform existing advanced classifiers in experiments, and its effectiveness has been verified by accuracy, precision, recall rate, and area under the ROC curve (OC-ROC). In addition, to accommodate computing resource constraints, we propose an optimized version of Deep-AvPIDen (DS), which utilizes Deep separation convolution technology to significantly reduce computing resource consumption. Through the online application platform, researchers can efficiently classify antiviral proteins and discover new AVPs. Future research could further optimize the model's computational efficiency, handle larger data sets, and expand its potential for biomedical problems such as drug combination prediction and new drug discovery.

FormulationBCS: A Machine Learning Platform Based on Diverse Molecular Representations for Biopharmaceutical Classification System (BCS) Class Prediction.

The Biopharmaceutics Classification System (BCS) has facilitated biowaivers and played a significant role in enhancing drug regulation and development efficiency. However, the productivity of measuring the key discriminative properties of BCS, solubility and permeability, still requires improvement, limiting high-throughput applications of BCS, which is essential for evaluating drug candidate developability and guiding formulation decisions in the early stages of drug development. In recent years, advancements in machine learning (ML) and molecular characterization have revealed the potential of quantitative structure-performance relationships (QSPR) for rapid and accurate in silico BCS classification. The present study aims to develop a web platform for high-throughput BCS classification based on high-performance ML models. Initially, four data sets of BCS-related molecular properties: log S, log P, log D, and log Papp were curated. Subsequently, 6 ML algorithms or deep learning frameworks were employed to construct models, with diverse molecular representations ranging from one-dimensional molecular fingerprints, descriptors, and molecular graphs to three-dimensional molecular spatial coordinates. By comparing different combinations of molecular representations and learning algorithms, LightGBM exhibited excellent performance in solubility prediction, with an R2 of 0.84; AttentiveFP outperformed others in permeability prediction, with R2 values of 0.96 and 0.76 for log P and log D, respectively; and XGBoost was the most accurate for log Papp prediction, with an R2 of 0.71. When externally validated on a marketed drug BCS category data set, the best-performing models achieved classification accuracies of over 77 and 73% for solubility and permeability, respectively. Finally, the well-trained models were embedded into the first ML-based BCS class prediction web platform (x f), enabling pharmaceutical scientists to quickly determine the BCS category of candidate drugs, which will aid in the high-throughput BCS assessment for candidate drugs during the preformulation stage, thereby promoting reduced risk and enhanced efficiency in drug development and regulation.

Leveraging Mendelian randomization to inform drug discovery and development for ischemic stroke.

Discovery and development of efficacious and safe pharmacological therapies is fraught with challenges. As proteins constitute the majority of drug targets and are encoded by genes, naturally occurring genetic variation within populations can provide valuable insights to inform drug discovery and development efforts. The drug target Mendelian randomization (MR) paradigm leverages these principles to investigate the causal effects of drug targets in humans. This review examines the application of drug target MR in informing the efficacy and development of therapeutics for ischemic stroke prevention and treatment. We consider applications of MR for existing and novel treatment strategies, including targeting blood pressure, lipid metabolism, coagulation, inflammation and glycemic control. Several of these genetically supported targets are under evaluation in late-stage clinical trials. Methodological limitations of drug target MR are addressed, followed by an outline of future research directions. We anticipate that careful application of drug target MR will enhance the efficiency of drug development for ischemic stroke, consequently accelerating the delivery of effective medications to patients.

Artificial intelligence-enabled social media listening to inform early patient-focused drug development: perspectives on approaches and strategies.

This article examines the opportunities and benefits of artificial intelligence (AI)-enabled social media listening (SML) in assisting successful patient-focused drug development (PFDD). PFDD aims to incorporate the patient perspective to improve the quality, relevance, safety, and efficiency of drug development and evaluation. Gathering patient perspectives to support PFDD is aided by the participation of patient groups in communicating their treatment experiences, needs, preferences, and priorities through online platforms. SML is a method of gathering feedback directly from patients; however, distilling the quantity of data into actionable insights is challenging. AI-enabled methods, such as natural language processing (NLP), can facilitate data processing from SML studies. Herein, we describe a novel, trainable, AI-enabled, SML workflow that classifies posts made by patients or caregivers and uses NLP to provide data on their experiences. Our approach is an iterative process that balances human expert-led milestones and AI-enabled processes to support data preprocessing, patient and caregiver classification, and NLP methods to produce qualitative data. We explored the applicability of this workflow in 2 studies: 1 in patients with head and neck cancers and another in patients with esophageal cancer. Continuous refinement of AI-enabled algorithms was essential for collecting accurate and valuable results. This approach and workflow contribute to the establishment of well-defined standards of SML studies and advance the methodologic quality and rigor of researchers contributing to, conducting, and evaluating SML studies in a PFDD context.

Employing Evolutionary Computing and Hybrid Artificial Neural Networks to Improve Drug Discovery

An ultimate relevant translational scientific endeavour that contributes to human vulnerability and happiness might be the creation and advancement of medications. Fast drug discovery procedures necessitate using contemporary computational approaches to tackle pharma data's complexities and high complexity. By combining Evolutionary Computing with Hybrid Artificial Neural Networks (EC-HANNs), this research introduces a novel strategy for optimizing and speeding up the drug development process. To handle various drug-target relations, predict the efficacy of compounds, and discover more accurate potential medications, the proposed model employs EC to evolve the computational construction and hyperparameters of HANNs in real-time. This innovation makes the model a flexible framework. Through repeated refinement of EC- HANN designs, the EC Module uses Particle Swarm Optimization (PSO) to choose the best configuration for individual drug discovery tasks. The model's distinctive feature is that it dynamically adjusts the network configuration to match the details of each drug discovery activity by using PSO to optimize the HANN's architecture and hyperparameters. Regarding sequential biological data, the HANN Module employs Convolutional Neural Networks (CNNs) to glean features for pattern recognition over time. By extracting high-level spatial information from genetic data, CNN can better identify prospective medication candidates. When tested on several benchmark datasets, the proposed framework demonstrates superior performance over traditional neural networks across the board regarding prediction accuracy, convergence speed, and model durability. As a powerful tool, this hybrid approach can simplify drug discovery, leading to more efficient drug development.

Advancements in Virtual Bioequivalence: A Systematic Review of Computational Methods and Regulatory Perspectives in the Pharmaceutical Industry.

The rise of virtual bioequivalence studies has transformed the pharmaceutical landscape, enabling more efficient drug development processes. This systematic review aims to explore advancements in physiologically based pharmacokinetic (PBPK) modeling, its regulatory implications, and its role in achieving virtual bioequivalence, particularly for complex drug formulations. We conducted a systematic review of clinical trials using computational methods, particularly PBPK modeling, to carry out bioequivalence assessments. Eligibility criteria are emphasized during in silico modeling and pharmacokinetic simulations. Comprehensive literature searches were performed across databases such as PubMed, Scopus, and the Cochrane Library. A search strategy using key terms and Boolean operators ensured that extensive coverage was achieved. We adhered to the PRISMA guidelines in regard to the study selection, data extraction, and quality assessment, focusing on key characteristics, methodologies, outcomes, and regulatory perspectives from the FDA and EMA. Our findings indicate that PBPK modeling significantly enhances the prediction of pharmacokinetic profiles, optimizing dosing regimens, while minimizing the need for extensive clinical trials. Regulatory agencies have recognized this utility, with the FDA and EMA developing frameworks to integrate in silico methods into drug evaluations. However, challenges such as study heterogeneity and publication bias may limit the generalizability of the results. This review highlights the critical need for standardized protocols and robust regulatory guidelines to facilitate the integration of virtual bioequivalence methodologies into pharmaceutical practices. By embracing these advancements, the pharmaceutical industry can improve drug development efficiency and patient outcomes, paving the way for innovative therapeutic solutions. Continued research and adaptive regulatory frameworks will be essential in navigating this evolving field.

Quinoline as a Privileged Structure: A Recent Update on Synthesis and Biological Activities.

Among heterocyclic compounds, quinoline is one of the best ubiquitous heterocyclic rings for medicinal chemistry purposes. Quinoline appears to be a powerful chemical structure to develop new drug entities. The quinoline derivatives own a wide array of biological activities such as anticancer, antimalarial, antimicrobial, anti-inflammatory, anti-leishmanial, etc. Because of the wide spectrum of bioactivities, the scientific communities are still looking for more efficient synthetic routes to form quinoline derivatives. Therefore, the primary focus of this review is to provide a thorough and inclusive, updated report on quinoline analogs that may pave the way for more efficient drug development.

A combined treatment with Ursolic acid and Solasodine inhibits colorectal cancer progression through the AKT1/ERK1/2-GSK-3β-β-catenin axis

BackgroundConventional chemotherapy medications are inadequate for managing the primary or acquired drug resistance, high toxicity, and adverse effects of colorectal cancer (CRC) treatment. Ursolic acid (UA) and Solasodine (Sol) are natural compounds found in a wide variety of traditional medicinal plants, as well as in many fruits and vegetables, such as Actinidia arguta (Sieb. & Zucc) Planch and Solanum nigrum L.. These compounds exhibit significant anti-tumor activity. Recent investigations have demonstrated that a combination strategy using natural products exhibits greater potential in CRC treatment compared to a single-drug strategy. PurposeThis study aimed to elucidate the potential of UA-Sol synergy against CRC and to investigate the mechanism of action involved in inducing apoptosis and inhibiting metastasis through the AKT1/ERK1/2-GSK-3β-β-catenin axis. MethodsThe optimal ratio of UA-Sol and its synergistic effects were explored using an MTT assay combined with the technique of Chou Talalay. The effects of UA-Sol on the apoptosis, autophagy, and metastasis of CRC cells were assessed using Annexin V-FITC/PI, TUNEL, Immunofluorescence, Wound healing, Transwell migration, and western blotting. The core mechanism of action of UA-Sol against CRC was investigated employing network pharmacology prediction combined with CETSA and plasmid transfection. Finally, in vivo validation was conducted using mouse xenograft tumor and lung metastasis models. ResultsThe combination of UA and Sol synergistically inhibited CRC cell viability at a molar ratio of 6:24. UA-Sol induced the expression of pro-apoptotic and autophagy genes such as Bax/Bcl-2 and LC3, ultimately leading to apoptosis and autophagy in CRC cells in vitro. In addition, this combination inhibited MMP-9 and promoted the expression of the adhesion protein E-cadherin, thereby inhibiting CRC cell metastasis. Mechanistically, UA-Sol regulated the expression of a downstream protein GSK-3β by targeting AKT1 and ERK1/2 inhibition. This induced a cross-talk between the MAPK cascade pathway and the PI3K/AKT pathway, thereby inhibiting the nuclear translocation of β-catenin and participating in the regulation of CRC cell processes. ConclusionUA-Sol inhibited the AKT1/ERK1/2-GSK-3β-β-catenin axis to induce apoptosis, autophagy and anti-metastasis by targeting AKT1 and ERK1/2 inhibition. This dual-target drug combination strategy provides promising insights into the development of novel, safe, and efficient drugs for the treatment of CRC.

Reevaluating safety pharmacology respiratory studies within the ICH S7A core battery: A multi-company evaluation of preclinical utility and clinical translation

Optimization of ICH safety guideline studies for inclusion into regulatory submissions is critical for resource conservation, animal use reduction, and efficient drug development. The ICH S7A guidance for Safety Pharmacology (SP) studies adopted in 2001 identified the core battery of studies to evaluate the acute safety of putative pharmaceutical molecules prior to First in Human (FIH) trials. To assess the utility of respiratory studies in predicting clinical AE's, seven pharmaceutical companies pooled preclinical and clinical respiratory findings. A large database of novel molecules included all relevant data from standard S7A respiratory (n = 459) and FIH studies (n = 309). The data were analyzed with respect to the progression of these molecules, clinical adverse event reporting of these same molecules, and achieved exposures. These S7A respiratory assay findings had no impact on compound progression, and only 12 of 309 drug candidates were ‘positive’ preclinically and reported a respiratory-related AE in clinical trials (i.e. cough, dyspnea, etc.), an overall incidence rate of 3.9%. Contingency tables/statistics support a lack of concordance of these preclinical assays. Overall, our extensive analysis clearly indicated that the preclinical respiratory assay fails to provide any prognostic value for detecting clinically relevant respiratory adverse events.

Anion exchange-HPLC method for evaluating the encapsulation efficiency of mRNA-loaded lipid nanoparticles using analytical quality by design

Lipid nanoparticles (LNPs) are emerging nucleic acid delivery systems in the development of mRNA therapeutics such as the severe acute respiratory syndrome coronavirus 2 vaccines. However, a suitable analytical method for evaluating the encapsulation efficiency (EE) of the LNPs is required to ensure drug efficacy, as current analytical methods exhibit throughput issues and require long analysis times. Hence, we developed and validated an anion-exchange HPLC method using Analytical Quality by Design. Three critical method parameters (CMPs) were identified using risk assessment and Design of Experiments: column temperature, flow rate, and sodium perchlorate concentration. The CMPs were optimized using Face-Centered Central Composite Design. The discriminating power of the optimized HPLC method and RiboGreen assay was comparable. The main advantage of this method is that LNPs can be directly injected into the HPLC system without bursting the LNPs loaded with encapsulated poly(A). The optimized HPLC method was validated as robust, high-throughput, and sufficiently sensitive according to the ICH Q2 guidelines. We believe our findings could promote efficient LNPs-based drug development.

Revolutionizing Drug Discovery: The Impact of Distinct Designs and Biosensor Integration in Microfluidics-Based Organ-on-a-Chip Technology.

Traditional drug development is a long and expensive process with high rates of failure. This has prompted the pharmaceutical industry to seek more efficient drug development frameworks, driving the emergence of organ-on-a-chip (OOC) based on microfluidic technologies. Unlike traditional animal experiments, OOC systems provide a more accurate simulation of human organ microenvironments and physiological responses, therefore offering a cost-effective and efficient platform for biomedical research, particularly in the development of new medicines. Additionally, OOC systems enable quick and real-time analysis, high-throughput experimentation, and automation. These advantages have shown significant promise in enhancing the drug development process. The success of an OOC system hinges on the integration of specific designs, manufacturing techniques, and biosensors to meet the need for integrated multiparameter datasets. This review focuses on the manufacturing, design, sensing systems, and applications of OOC systems, highlighting their design and sensing capabilities, as well as the technical challenges they currently face.

Predicting drug-target interactions by measuring confidence with consistent causal neighborhood interventions

Predicting drug-target interactions (DTI) is a crucial stage in drug discovery and development. Understanding the interaction between drugs and targets is essential for pinpointing the specific relationship between drug molecules and targets, akin to solving a link prediction problem using information technology. While knowledge graph (KG) and knowledge graph embedding (KGE) methods have been rapid advancements and demonstrated impressive performance in drug discovery, they often lack authenticity and accuracy in identifying DTI. This leads to increased misjudgment rates and reduced efficiency in drug development. To address these challenges, our focus lies in refining the accuracy of DTI prediction models through KGE, with a specific emphasis on causal intervention confidence measures (CI). These measures aim to assess triplet scores, enhancing the precision of the predictions. Comparative experiments conducted on three datasets and utilizing 9 KGE models reveal that our proposed confidence measure approach via causal intervention, significantly improves the accuracy of DTI link prediction compared to traditional approaches. Furthermore, our experimental analysis delves deeper into the embedding of intervention values, offering valuable insights for guiding the design and development of subsequent drug development experiments. As a result, our predicted outcomes serve as valuable guidance in the pursuit of more efficient drug development processes.

Next-generation pediatric care: nanotechnology-based and AI-driven solutions for cardiovascular, respiratory, and gastrointestinal disorders.

Global pediatric healthcare reveals significant morbidity and mortality rates linked to respiratory, cardiac, and gastrointestinal disorders in children and newborns, mostly due to the complexity of therapeutic management in pediatrics and neonatology, owing to the lack of suitable dosage forms for these patients, often rendering them "therapeutic orphans". The development and application of pediatric drug formulations encounter numerous challenges, including physiological heterogeneity within age groups, limited profitability for the pharmaceutical industry, and ethical and clinical constraints. Many drugs are used unlicensed or off-label, posing a high risk of toxicity and reduced efficacy. Despite these circumstances, some regulatory changes are being performed, thus thrusting research innovation in this field. Up-to-date peer-reviewed journal articles, books, government and institutional reports, data repositories and databases were used as main data sources. Among the main strategies proposed to address the current pediatric care situation, nanotechnology is specially promising for pediatric respiratory diseases since they offer a non-invasive, versatile, tunable, site-specific drug release. Tissue engineering is in the spotlight as strategy to address pediatric cardiac diseases, together with theragnostic systems. The integration of nanotechnology and theragnostic stands poised to refine and propel nanomedicine approaches, ushering in an era of innovative and personalized drug delivery for pediatric patients. Finally, the intersection of drug repurposing and artificial intelligence tools in pediatric healthcare holds great potential. This promises not only to enhance efficiency in drug development in general, but also in the pediatric field, hopefully boosting clinical trials for this population. Despite the long road ahead, the deepening of nanotechnology, the evolution of tissue engineering, and the combination of traditional techniques with artificial intelligence are the most recently reported strategies in the specific field of pediatric therapeutics.

Mini-colons predict drug toxicity in vitro

Mitrofanova etal.1 engineer a human colonic invitro model capable of producing an intestinal mucus barrier, with potential applications for predicting drug-induced gastrointestinal toxicity. This improved system paves the way for more accurate and efficient drug development processes.

Machine learning identifies prognostic subtypes of the tumor microenvironment of NSCLC

The tumor microenvironment (TME) plays a fundamental role in tumorigenesis, tumor progression, and anti-cancer immunity potential of emerging cancer therapeutics. Understanding inter-patient TME heterogeneity, however, remains a challenge to efficient drug development. This article applies recent advances in machine learning (ML) for survival analysis to a retrospective study of NSCLC patients who received definitive surgical resection and immune pathology following surgery. ML methods are compared for their effectiveness in identifying prognostic subtypes. Six survival models, including Cox regression and five survival machine learning methods, were calibrated and applied to predict survival for NSCLC patients based on PD-L1 expression, CD3 expression, and ten baseline patient characteristics. Prognostic subregions of the biomarker space are delineated for each method using synthetic patient data augmentation and compared between models for overall survival concordance. A total of 423 NSCLC patients (46% female; median age [inter quantile range]: 67 [60–73]) treated with definite surgical resection were included in the study. And 219 (52%) patients experienced events during the observation period consisting of a maximum follow-up of 10 years and median follow up 78 months. The random survival forest (RSF) achieved the highest predictive accuracy, with a C-index of 0.84. The resultant biomarker subtypes demonstrate that patients with high PD-L1 expression combined with low CD3 counts experience higher risk of death within five-years of surgical resection.

The Antifungal Potential of Niclosamide and Structurally Related Salicylanilides.

Human mycoses cover a diverse field of fungal diseases from skin disorders to systemic invasive infections and pose an increasing global health problem based on ineffective treatment options, the hampered development of new efficient drugs, and the emergence of resistant fungal strains. Niclosamide is currently applied for the treatment of worm infections. Its mechanisms of action, which include the suppression of mitochondrial oxidative phosphorylation (also known as mitochondrial uncoupling), among others, has led to a repurposing of this promising anthelmintic drug for the therapy of further human diseases such as cancer, diabetes, and microbial infections. Given the urgent need to develop new drugs against fungal infections, the considerable antifungal properties of niclosamide are highlighted in this review. Its chemical and pharmacological properties relevant for drug development are also briefly mentioned, and the described mitochondria-targeting mechanisms of action add to the current arsenal of approved antifungal drugs. In addition, the activities of further salicylanilide-based niclosamide analogs against fungal pathogens, including agents applied in veterinary medicine for many years, are described and discussed for their feasibility as new antifungals for humans. Preliminary structure-activity relationships are determined and discussed. Various salicylanilide derivatives with antifungal activities showed increased oral bioavailabilities when compared with niclosamide. The simple synthesis of salicylanilide-based drugs also vouchsafes a broad and cost-effective availability for poorer patient groups. Pertinent literature is covered until 2024.

AIGO-DTI: Predicting Drug-Target Interactions Based on Improved Drug Properties Combined with Adaptive Iterative Algorithms.

Artificial intelligence-based methods for predicting drug-target interactions (DTIs) aim to explore reliable drug candidate targets rapidly and cost-effectively to accelerate the drug development process. However, current methods are often limited by the topological regularities of drug molecules, making them difficult to generalize to a broader chemical space. Additionally, the use of similarity to measure DTI network links often introduces noise, leading to false DTI relationships and affecting the prediction accuracy. To address these issues, this study proposes an Adaptive Iterative Graph Optimization (AIGO)-DTI prediction framework. This framework integrates atomic cluster information and enhances molecular features through the design of functional group prompts and graph encoders, optimizing the construction of DTI association networks. Furthermore, the optimization of graph structure is transformed into a node similarity learning problem, utilizing multihead similarity metric functions to iteratively update the network structure to improve the quality of DTI information. Experimental results demonstrate the outstanding performance of AIGO-DTI on multiple public data sets and label reversal data sets. Case studies, molecular docking, and existing research validate its effectiveness and reliability. Overall, the method proposed in this study can construct comprehensive and reliable DTI association network information, providing new graphing and optimization strategies for DTI prediction, which contribute to efficient drug development and reduce target discovery costs.

Exploring optimal drug targets through subtractive proteomics analysis and pangenomic insights for tailored drug design in tuberculosis

Tuberculosis (TB), caused by Mycobacterium tuberculosis, ranks among the top causes of global human mortality, as reported by the World Health Organization’s 2022 TB report. The prevalence of M. tuberculosis strains that are multiple and extensive-drug resistant represents a significant barrier to TB eradication. Fortunately, having many completely sequenced M. tuberculosis genomes available has made it possible to investigate the species pangenome, conduct a pan-phylogenetic investigation, and find potential new drug targets. The 442 complete genome dataset was used to estimate the pangenome of M. tuberculosis. This study involved phylogenomic classification and in-depth analyses. Sequential filters were applied to the conserved core genome containing 2754 proteins. These filters assessed non-human homology, virulence, essentiality, physiochemical properties, and pathway analysis. Through these intensive filtering approaches, promising broad-spectrum therapeutic targets were identified. These targets were docked with FDA-approved compounds readily available on the ZINC database. Selected highly ranked ligands with inhibitory potential include dihydroergotamine and abiraterone acetate. The effectiveness of the ligands has been supported by molecular dynamics simulation of the ligand–protein complexes, instilling optimism that the identified lead compounds may serve as a robust basis for the development of safe and efficient drugs for TB treatment, subject to further lead optimization and subsequent experimental validation.

Erianin inhibits the progression of triple-negative breast cancer by suppressing SRC-mediated cholesterol metabolism

Triple-negative breast cancer (TNBC) is highly malignant and lacks effective biotherapeutic targets. The development of efficient anticancer drugs with low toxicity and few side effects is a hotspot in TNBC treatment research. Although erianin is known to have potent antitumor activity, its regulatory mechanism and target in TNBC have not been fully elucidated, hampering further drug development. This study showed that erianin can significantly inhibit TNBC cell proliferation and migration, promote cell apoptosis, and inhibit the growth of transplanted tumors in mice. Mechanistically, through network pharmacology analysis, molecular docking and cellular thermal shift assays, we preliminarily identified SRC as the cellular target of erianin. Erianin potently inhibited the expression of SRC, which mediated the anticancer effect of erianin in TNBC. Moreover, erianin can downregulate the expression of genes related to cholesterol synthesis and uptake by targeting SRC, interfering with cholesterol levels in TNBC, thereby inhibiting the progression of TNBC in vivo and in vitro. Taken together, our results suggest that erianin may inhibit the progression of TNBC by suppressing SRC-mediated cholesterol metabolism, and erianin has the great potential to be an effective treatment for TNBC patients.

Prediction of Drug-Target Affinity Using Attention Neural Network.

Studying drug-target interactions (DTIs) is the foundational and crucial phase in drug discovery. Biochemical experiments, while being the most reliable method for determining drug-target affinity (DTA), are time-consuming and costly, making it challenging to meet the current demands for swift and efficient drug development. Consequently, computational DTA prediction methods have emerged as indispensable tools for this research. In this article, we propose a novel deep learning algorithm named GRA-DTA, for DTA prediction. Specifically, we introduce Bidirectional Gated Recurrent Unit (BiGRU) combined with a soft attention mechanism to learn target representations. We employ Graph Sample and Aggregate (GraphSAGE) to learn drug representation, especially to distinguish the different features of drug and target representations and their dimensional contributions. We merge drug and target representations by an attention neural network (ANN) to learn drug-target pair representations, which are fed into fully connected layers to yield predictive DTA. The experimental results showed that GRA-DTA achieved mean squared error of 0.142 and 0.225 and concordance index reached 0.897 and 0.890 on the benchmark datasets KIBA and Davis, respectively, surpassing the most state-of-the-art DTA prediction algorithms.