Molecular Structural Information Research Articles

CIDGMed: Causal Inference-Driven Medication Recommendation with enhanced dual-granularity learning

Medication recommendation aims to integrate patients’ long-term health records to provide accurate and safe medication combinations for specific health states. Existing methods often fail to deeply explore the true causal relationships between diseases/procedures and medications, resulting in biased recommendations. Additionally, in medication representation learning, the relationships between information at different granularities of medications—coarse-grained (medication itself) and fine-grained (molecular level)—are not effectively integrated, leading to biases in representation learning. To address these limitations, we propose the Causal Inference-driven Dual-Granularity Medication Recommendation method (CIDGMed). Our approach leverages causal inference to uncover the relationships between diseases/procedures and medications, thereby enhancing the rationality and interpretability of recommendations. By integrating coarse-grained medication effects with fine-grained molecular structure information, CIDGMed provides a comprehensive representation of medications. Additionally, we employ a bias correction model during the prediction phase to further refine recommendations, ensuring both accuracy and safety. Through extensive experiments, CIDGMed significantly outperforms current state-of-the-art models across multiple metrics, achieving a 2.54% increase in accuracy, a 3.65% reduction in side effects, and a 39.42% improvement in time efficiency. Additionally, we demonstrate the rationale of CIDGMed through a case study.

Integrated Knowledge Graph and Drug Molecular Graph Fusion via Adversarial Networks for Drug-Drug Interaction Prediction.

The Co-administration of multiple drugs can enhance the efficacy of disease treatment by reducing drug resistance and side effects. However, it also raises the risk of adverse drug interactions, presenting a challenging problem in healthcare. Various approaches have been developed to predict drug-drug interactions (DDIs) by leveraging both knowledge graphs and drug attribute information. While these methods have shown promise, they often fail to effectively capture correlations between biomedical information in the knowledge graph and drug properties. This work introduces a novel end-to-end DDI predictor framework based on generative adversarial networks. This framework utilizes a message-passing neural network to capture molecular structure information while employing the knowledge-aware graph attention network to capture the representation of drugs in the knowledge graph through considering the importance of different multihop neighbor nodes and relationships. The dual generative adversarial networks employ two generators and two discriminators in knowledge graph embedding and molecular topology embedding for adversarial training to capture the interrelations and complementary knowledge between molecular structure information and semantic information from the knowledge graph. comprehensive experiments have demonstrated that the proposed method outperforms state-of-the-art algorithms in binary classification, with improvements of 1.0% in accuracy, 0.45% in area under the receiver operating characteristic curve (AUC), 0.24% in area under the precision-recall curve (AUPR), and 0.98% in F1 score. Furthermore, for multiclass classification tasks, improvements were observed across various evaluation metrics, including 0.9% in accuracy, 0.25% in macro precision, 0.2% in macro recall, and 0.28% in macro F1. Additionally, ablation studies were conducted to showcase the effectiveness and robustness of our method in DDI prediction tasks.

A Knowledge-Driven Self-Supervised Approach for Molecular Generation.

Due to the great successes of Graph Neural Networks (GNN) in numerous fields, growing research interests have been devoted to applying GNN to molecular learning tasks. The molecule structure can be naturally represented as graphs where atoms and bonds refer to nodes and edges respectively. However, the atoms are not haphazardly stacked together but combined into various spatial geometries. Meanwhile, since chemical reactions mainly occur in substructures such as functional groups, the substructure plays a decisive role in the molecule's properties. Therefore, directly applying GNN to molecular representation learning could ignore the molecular spatial structure and the substructure properties which in turn degrades the performance of downstream tasks. In this paper, we propose Knowledge-Driven Self-Supervised Model for Molecular Representation Learning (KSMRL) to address above problems. The KSMRL consists of two major pathways: (1) the Spatial Information (SI) based pathway which preserves the spatial information of molecular structure, (2) the Subgraph Constraint (SC) based pathway which retains the properties of substructures into the molecular representation. In this manner, both the atomic level and substructure level information can be included in modeling. According to the experimental results on multiple datasets, the proposed KSMRL can generate discriminative molecular representations. In molecular generation tasks, KSMRL combined with Autoregressive Flow (AF) models or Discrete Flow (DF) models outperforms the state-of-the-art baselines over all datasets. In addition, we demonstrate the effectiveness of KSMRL with property optimization experiments. To indicate the ability of predicting specified potential Drug-Target Interactions (DTIs), a case study for discriminating the interactions between molecule generated by KSMRL and targets is also given.

Metabolic Pathways of Acylcarnitine Synthesis.

Acylcarnitines are important markers in metabolic studies of many diseases, including metabolic, cardiovascular, and neurological disorders. We reviewed analytical methods for analyzing acylcarnitines with respect to the available molecular structural information, the technical limitations of legacy methods, and the potential of new mass spectrometry-based techniques to provide new information on metabolite structure. We summarized the nomenclature of acylcarnitines based on historical common names and common abbreviations, and we propose the use of systematic abbreviations derived from the shorthand notation for lipid structures. The transition to systematic nomenclature will facilitate acylcarnitine annotation, reporting, and standardization in metabolomics. We have reviewed the metabolic origins of acylcarnitines important for the biological interpretation of human metabolomic profiles. We identified neglected isomers of acylcarnitines and summarized the metabolic pathways involved in the synthesis and degradation of acylcarnitines, including branched-chain lipids and amino acids. We reviewed the primary literature, mapped the metabolic transformations of acyl-CoAs to acylcarnitines, and created a freely available WikiPathway WP5423 to help researchers navigate the acylcarnitine field. The WikiPathway was curated, metabolites and metabolic reactions were annotated, and references were included. We also provide a table for conversion between common names and abbreviations and systematic abbreviations linked to the LIPID MAPS or Human Metabolome Database.

DRN-CDR: A cancer drug response prediction model using multi-omics and drug features

Cancer drug response (CDR) prediction is an important area of research that aims to personalize cancer therapy, optimizing treatment plans for maximum effectiveness while minimizing potential negative effects. Despite the advancements in Deep learning techniques, the effective integration of multi-omics data for drug response prediction remains challenging. In this paper, a regression method using Deep ResNet for CDR (DRN-CDR) prediction is proposed. We aim to explore the potential of considering sole cancer genes in drug response prediction. Here the multi-omics data such as gene expressions, mutation data, and methylation data along with the molecular structural information of drugs were integrated to predict the IC50 values of drugs. Drug features are extracted by employing a Uniform Graph Convolution Network, while Cell line features are extracted using a combination of Convolutional Neural Network and Fully Connected Networks. These features are then concatenated and fed into a deep ResNet for the prediction of IC50 values between Drug – Cell line pairs. The proposed method yielded higher Pearson’s correlation coefficient (rp) of 0.7938 with lowest Root Mean Squared Error (RMSE) value of 0.92 when compared with similar methods of tCNNS, MOLI, DeepCDR, TGSA, NIHGCN, DeepTTA, GraTransDRP and TSGCNN. Further, when the model is extended to a classification problem to categorize drugs as sensitive or resistant, we achieved AUC and AUPR measures of 0.7623 and 0.7691, respectively. The drugs such as Tivozanib, SNX-2112, CGP-60474, PHA-665752, Foretinib etc., exhibited low median IC50 values and were found to be effective anti-cancer drugs. The case studies with different TCGA cancer types also revealed the effectiveness of SNX-2112, CGP-60474, Foretinib, Cisplatin, Vinblastine etc. This consistent pattern strongly suggests the effectiveness of the model in predicting CDR.

From the TOP: Formation, recognition and resolution of topoisomerase DNA protein crosslinks

Since the report of “DNA untwisting” activity in 1972, ∼50 years of research has revealed seven topoisomerases in humans (TOP1, TOP1mt, TOP2α, TOP2β, TOP3α, TOP3β and Spo11). These conserved regulators of DNA topology catalyze controlled breakage to the DNA backbone to relieve the torsional stress that accumulates during essential DNA transactions including DNA replication, transcription, and DNA repair. Each topoisomerase-catalyzed reaction involves the formation of a topoisomerase cleavage complex (TOPcc), a covalent protein-DNA reaction intermediate formed between the DNA phosphodiester backbone and a topoisomerase catalytic tyrosine residue. A variety of perturbations to topoisomerase reaction cycles can trigger failure of the enzyme to re-ligate the broken DNA strand(s), thereby generating topoisomerase DNA-protein crosslinks (TOP-DPC). TOP-DPCs pose unique threats to genomic integrity. These complex lesions are comprised of structurally diverse protein components covalently linked to genomic DNA, which are bulky DNA adducts that can directly impact progression of the transcription and DNA replication apparatus. A variety of genome maintenance pathways have evolved to recognize and resolve TOP-DPCs. Eukaryotic cells harbor tyrosyl DNA phosphodiesterases (TDPs) that directly reverse 3′-phosphotyrosyl (TDP1) and 5′-phoshotyrosyl (TDP2) protein-DNA linkages. The broad specificity Mre11-Rad50-Nbs1 and APE2 nucleases are also critical for mitigating topoisomerase-generated DNA damage. These DNA-protein crosslink metabolizing enzymes are further enabled by proteolytic degradation, with the proteasome, Spartan, GCNA, Ddi2, and FAM111A proteases implicated thus far. Strategies to target, unfold, and degrade the protein component of TOP-DPCs have evolved as well. Here we survey mechanisms for addressing Topoisomerase 1 (TOP1) and Topoisomerase 2 (TOP2) DPCs, highlighting systems for which molecular structure information has illuminated function of these critical DNA damage response pathways.

A Multibranch Neural Network for Drug-Target Affinity Prediction Using Similarity Information.

Predicting drug-target affinity (DTA) is beneficial for accelerating drug discovery. In recent years, graph structure-based deep learning models have garnered significant attention in this field. However, these models typically handle drug or target protein in isolation and only extract the molecular structure information on the drug or protein itself. To address this limitation, existing network-based models represent drug-target interactions or affinities as a knowledge graph to capture the interaction information. In this study, we propose a novel solution. Specifically, we introduce drug similarity information and protein similarity information into the field of DTA prediction. Moreover, we propose a network framework that autonomously extracts similarity information, avoiding reliance on knowledge graphs. Based on this framework, we design a multibranch neural network called GASI-DTA. This network integrates similarity information, sequence information, and molecular structure information. Comprehensive experimental results conducted on two benchmark data sets and three cold-start scenarios demonstrate that our model outperforms state-of-the-art graph structure-based methods in nearly all metrics. Furthermore, it exhibits significant advantages over existing network-based models, outperforming the best of them in the majority of metrics. Our study's code and data are openly accessible at http://github.com/XiaoLin-Yang-S/GASI-DTA.

3DSGIMD: An accurate and interpretable molecular property prediction method using 3D spatial graph focusing network and structure-based feature fusion

A comprehensive representation of molecular structure is essential for establishing accurate and reliable molecular property prediction models. However, fully extracting and learning intrinsic molecular structure information, especially spatial structure features, remains a challenging task, leading that many molecular property prediction models still have no enough accuracy for the real application. In this study, we developed an innovative and interpretable deep learning method, termed 3DSGIMD, which predicted the molecular properties by integrating and learning the spatial structure and substructure information of molecules at multiple levels, and generated the focusing weights by aggregating spatial and adjacency information of molecules to improve understanding of prediction results. We evaluated the model on 10 public datasets and 14 cell-based phenotypic screening datasets. Extensive experimental results indicated that 3DSGIMD achieved superior or comparable predictive performance compared with some existing models, and the individually designed components contributed significantly to the advanced performance of the model. In addition, we also provided insight into the interpretability of our model via visualizing the focusing weights and perturbation analysis, and the results showed that 3DSGIMD can pinpoint crucial local structures and bits of molecular descriptors associated with the predicted properties. In summary, 3DSGIMD is a competitive molecular property prediction method that holds the potential to aid drug design and optimization.

Classification Model of Pesticide Toxicity in Americamysis bahia Based on Quantum Chemical Descriptors.

A set of quantum chemical descriptors (molecular polarization, heat capacity, entropy, Mulliken net charge of the most positive hydrogen atom, APT charge of the most negative atom and APT charge of the most positive atom with hydrogen summed into heavy atoms) was successfully used to establish the classification models for the toxicity pLC50 of pesticides in Americamysis bahia. The optimal random forest model (Class Model A) yielded predictive accuracy of 100% (training set of 217 pesticides), 95.8% (test set of 72 pesticides) and 99.0% (total set of 289 pesticides), which were very satisfactory, compared with previous classification models reported for the toxicity of compounds in aquatic organisms. Therefore, it is reasonable to apply the quantum chemical descriptors associated with molecular structural information on molecular bulk, chemical reactivity and weak interactions, to develop classification models for the toxicity pLC50 of pesticides in A. bahia.

Calculation of the Structure and Reaction Potential Barrier of the Decomposition Products of SF6/N2 Mixed Gases

Abstract The greenhouse effect caused by SF6 gas in the power system cannot be ignored, and N2 has enormous industrial potential to replace pure SF6 by mixing with SF6 due to its advantages of low environmental pollution, low price, and stable chemical performance. This article combines density functional theory (DFT) and transition state theory (TST) to calculate the molecular structure and vibrational frequency information of various reactants, products, and transition states involved in SF6/N2 gas mixtures containing water and oxygen. The single point energy of each component is calculated using the CCSD (T)/cc pvtz theory level, and all reaction barrier energies are obtained. The calculated molecular structure and vibration frequency results are in good agreement with the experimental data of NIST, proving the reliability of the calculation method in this paper. The difficulty of the reaction can be determined by obtaining all the reaction barrier energies. It was found that SF5 molecules are unstable and prone to decomposition. SF4 is very important in the SF6/N2 reaction system, and further decomposition will produce a large number of stable by-products. The calculation method in this article helps to deeply understand the decomposition reaction mechanism of SF6/N2 mixed gas, and is of great significance for the research of decomposition experiments.

Analysis of the effect of cations on protein conformational stability using solid-state nanopores.

The conformation of proteins is closely related to their biological functions, and it is affected by many factors, including the type of cations in solution. However, it is difficult to detect the conformational changes of a protein in situ. As a single-molecule sensing technology, nanopores can convert molecular structural information into analyzable current signals within a reasonable time range. Herein, we detect and analyze the effects of two different types of monovalent cations (Na+ and Li+) on a model protein bovine serum albumin (BSA) conformation using SiNx nanopores with different diameters. The quantitative analysis results show that the excluded volume of BSA in LiCl salt solutions is larger than the value in NaCl solution, indicating that Li+ is more prone to unfolding the proteins and making them unstable. This study demonstrated that nanopores enable the in situ detection of the structure of proteins at the single-molecule level and provide a new approach for the quantitative analysis of proteins.

A variational expectation-maximization framework for balanced multi-scale learning of protein and drug interactions

Protein functions are characterized by interactions with proteins, drugs, and other biomolecules. Understanding these interactions is essential for deciphering the molecular mechanisms underlying biological processes and developing new therapeutic strategies. Current computational methods mostly predict interactions based on either molecular network or structural information, without integrating them within a unified multi-scale framework. While a few multi-view learning methods are devoted to fusing the multi-scale information, these methods tend to rely intensively on a single scale and under-fitting the others, likely attributed to the imbalanced nature and inherent greediness of multi-scale learning. To alleviate the optimization imbalance, we present MUSE, a multi-scale representation learning framework based on a variant expectation maximization to optimize different scales in an alternating procedure over multiple iterations. This strategy efficiently fuses multi-scale information between atomic structure and molecular network scale through mutual supervision and iterative optimization. MUSE outperforms the current state-of-the-art models not only in molecular interaction (protein-protein, drug-protein, and drug-drug) tasks but also in protein interface prediction at the atomic structure scale. More importantly, the multi-scale learning framework shows potential for extension to other scales of computational drug discovery.

Molecular fingerprint-aided prediction of organic solute rejection in reverse osmosis and nanofiltration

Reverse osmosis and nanofiltration are used to purify feedwaters that contain a range of harmful organic solutes. The rejection of many of these solutes is poorly understood due to our limited ability to experimentally measure removal of any given compound. In this work, we present a machine learning approach that predicts organic solute rejection using molecular fingerprints that encode chemical structure features, such as functional groups and rings, into simple binary vectors. We trained machine learning models on a database of 1906 membrane rejection measurements including 228 organic compounds and 39 types of reverse osmosis and nanofiltration membranes. Three types of molecular fingerprint models (structural key, circular, and path based) were compared, and we observed that the Molecular Access System (MACCS) structural key had high performance (coefficient of determination of 0.87 with the testing set), fast calculation time due to its short bit-length, and easy interpretability. In addition to evaluating prediction performance, Shapley Additive Explanations (SHAP) analysis was implemented to gain a better molecular-scale understanding of membrane rejection, identifying molecular substructures that are important in determining their rejection. Overall, this work presents a method to predict the rejection of compounds that uses readily available molecular structure information and improves our ability to understand rejection mechanisms.

Irradiation multi-scale damage and interface effects of 3D braided carbon fiber/epoxy composites subjected to high dose γ-rays

The serious damage caused to 3D braided carbon fiber (CF)/epoxy (EP) composites in high-energy irradiation environments is a pressing research topic that has not yet been reported. This study analyzed the γ-irradiation multi-scale damage of the matrix, interfaces, and near-interface regions in 3D braided CF/EP composites with three different braiding angles (18°, 28°, 38°) at atomic, microscopic, and macroscopic scales. The molecular structure and atomic charge information alterations of epoxy matrix were characterized using soft X-ray absorption spectroscopy (sXAS). When the irradiation dose reached 1000 KGy, the compressive strength of composites decreased by 20.88 %, 18.07 %, and 16.60 %, respectively, with an increase in the braiding angle. Micro-CT observation, along with statistical computing, confirmed that irradiation causes interface damage and increases the defect volume of 3D braided composites. Nanoindentation was used to compare the modulus of carbon fiber, interface, near-interface regions and matrix in irradiated composites and the function of CF/resin interface as an irradiation defect capture site in the irradiation resistance of resin-matrix composites has been newly established. In addition, the mechanism behind the effect of braiding angle on macroscopic properties was also analyzed.

Functional and structural characterization of an IclR family transcription factor for the development of dicarboxylic acid biosensors.

Prokaryotic transcription factors (TFs) regulate gene expression in response to small molecules, thus representing promising candidates as versatile small molecule-detecting biosensors valuable for synthetic biology applications. The engineering of such biosensors requires thorough in vitro and in vivo characterization of TF ligand response as well as detailed molecular structure information. In this work, we functionally and structurally characterize the Pca regulon regulatory protein (PcaR) transcription factor belonging to the IclR transcription factor family. Here, we present in vitro functional analysis of the ligand profile of PcaR and the construction of genetic circuits for the characterization of PcaR as an in vivo biosensor in the model eukaryote Saccharomyces cerevisiae. We report the crystal structures of PcaR in the apo state and in complex with one of its ligands, succinate, which suggests the mechanism of dicarboxylic acid recognition by this transcription factor. This work contributes key structural and functional insights enabling the engineering of PcaR for dicarboxylic acid biosensors, in addition to providing more insights into the IclR family of regulators.

Conformation Influence of DNA on the Detection Signal through Solid-State Nanopores.

The detection and identification of nanoscale molecules are crucial, but traditional technology comes with a high cost and requires skilled operators. Solid-state nanopores are new powerful tools for discerning the three-dimensional shape and size of molecules, enabling the translation of molecular structural information into electric signals. Here, DNA molecules with different shapes were designed to explore the effects of electroosmotic forces (EOF), electrophoretic forces (EPF), and volume exclusion on electric signals within solid-state nanopores. Our results revealed that the electroosmotic force was the main driving force for single-stranded DNA (ssDNA), whereas double-stranded DNA (dsDNA) was primarily dominated by electrophoretic forces in nanopores. Moreover, dsDNA caused greater amplitude signals and moved faster through the nanopore due to its larger diameter and carrying more charges. Furthermore, at the same charge level and amount of bases, circular dsDNA exhibited a tighter structure compared to brush DNA, resulting in a shorter length. Consequently, circular dsDNA caused higher current-blocking amplitudes and faster passage speeds. The characterization approach based on nanopores allows researchers to get molecular information about size and shape in real time. These findings suggest that nanopore detection has the potential to streamline nanoscale characterization and analysis, potentially reducing both the cost and complexity.

GCARDTI: Drug–target interaction prediction based on a hybrid mechanism in drug SELFIES

AbstractThe prediction of the interaction between a drug and a target is the most critical issue in the fields of drug development and repurposing. However, there are still two challenges in current deep learning research: (i) the structural information of drug molecules is not fully explored in most drug target studies, and the previous drug SMILES does not correspond well to effective drug molecules and (ii) exploration of the potential relationship between drugs and targets is in need of improvement. In this work, we use a new and better representation of the effective molecular graph structure, SELFIES. We propose a hybrid mechanism framework based on convolutional neural network and graph attention network to capture multi‐view feature information of drug and target molecular structures, and we aim to enhance the ability to capture interaction sites between a drug and a target. In this study, our experiments using two different datasets show that the GCARDTI model outperforms a variety of different model algorithms on different metrics. We also demonstrate the accuracy of our model through two case studies.

Neural network-based prediction of auto-ignition temperature of ternary mixed liquids

Auto-ignition temperature (AIT) is one of the crucial exponents in the design of fire and explosion safety measures. Therefore, in this study, quantitative structure-property relationship approach was used to predict the AIT of ternary hybrid liquids based on molecular structure information. The optimal molecular descriptors were calculated and filtered using Mordred software. Twelve mixing rules were proposed for calculating molecular descriptors of mixtures. A prediction model for the AIT value of binary liquid mixtures was developed, validated and evaluated using a back propagation neural network (BPNN) and a one-dimensional convolutional neural network (1DCNN). The relative contribution and positive and negative correlations between individual molecular descriptors and AIT in the model were interpreted using the shapley additive explanations method. The results show that BPNN and 1DCNN models using mixing rule 1 have the best fitting ability, stability and prediction ability. The determination coefficient of the BPNN and 1DCNN models in the training set were 0.996 and 0.992, the root mean square errors were 3.613 °C and 5.284 °C, the mean absolute errors were 2.483 °C and 4.144 °C, the nash efficiency coefficient was 0.996 and 0.992, respectively, the willmott index was 0.999 and 0.998. and the values of the top three molecular descriptors of relative contribution, SssCH2, SsOH and SsCH3, were negatively correlated with the AIT values. The BPNN and 1DCNN models provide an accurate and reliable method for predicting ternary mixing liquid AIT.

G-K BertDTA: A graph representation learning and semantic embedding-based framework for drug-target affinity prediction

Developing new drugs is costly, time-consuming, and risky. Drug-target affinity (DTA), indicating the binding capability between drugs and target proteins, is a crucial indicator for drug development. Accurately predicting interaction strength between new drug-target pairs by analyzing previous experiments aids in screening potential drug molecules, repurposing them, and developing safe and effective medicines. Existing computational models for DTA prediction rely on strings or single-graph neural networks, lacking consideration of protein structure and molecular semantic information, leading to limited accuracy. Our experiments demonstrate that string-based methods may overlook protein conformations, causing a high root mean square error (RMSE) of 3.584 in affinity due to a lack of spatial context. Single graph networks also underperform on topology features, with a 6% lower confidence interval (CI) for activity classification. Absent semantic information also limits generalization across diverse compounds, resulting in 18% increment in RMSE and 5% in misclassifications within quantifications study, restricting potential drug discovery. To address these limitations, we propose G-K BertDTA, a novel framework for accurate DTA prediction incorporating protein features, molecular semantic features, and molecular structural information. In this proposed model, we represent drugs as graphs, with a GIN employed to learn the molecular topological information. For the extraction of protein structural features, we utilize a DenseNet architecture. A knowledge-based BERT semantic model is incorporated to obtain rich pre-trained semantic embeddings, thereby enhancing the feature information. We extensively evaluated our proposed approach on the publicly available benchmark datasets (i.e., KIBA and Davis), and experimental results demonstrate the promising performance of our method, which consistently outperforms previous state-of-the-art approaches. Code is available at https://github.com/AmbitYuki/G-K-BertDTA.

Attention-Based Learning for Predicting Drug-Drug Interactions in Knowledge Graph Embedding Based on Multisource Fusion Information

Drug combinations can reduce drug resistance and side effects and enable the improvement of disease treatment efficacy. Therefore, how to effectively identify drug-drug interactions (DDIs) is a challenging problem. Currently, there exist several approaches that leverage advanced representation learning and graph-based techniques for DDIs prediction. While these methods have demonstrated promising results, a limited number of approaches effectively utilize the potential of knowledge graphs (KGs), which provide information on drug attributes and multirelation among entities. In this work, we introduce a novel attention-based KGs representation learning framework. To encode drug SMILES sequence, a pretrained model is used, while molecular structure information is mapped as the initialization of nodes within the KG using a message-passing neural network. Additionally, the knowledge-aware graph attention network is employed to capture the drug and its topological neighbor representation in the KG representation module. To prevent the oversmoothing problem, the residual layer is used in the DDI prediction module. Comprehensive experiments on several datasets have demonstrated that the proposed method outperforms the state-of-the-art algorithms on the DDI prediction task across a range of evaluation metrics. It achieves an accuracy of 0.924 and an AUC of 0.9705 on the KEGG dataset and attains an ACC of 0.9777 and an AUC of 0.9959 on the OGB-biokg dataset. These experimental findings affirm that our approach is a dependable model for predicting the association of drugs.