Drug-drug interactions prediction based on deep learning and knowledge graph: A review.

Drug-drug interactions (DDIs) can produce unpredictable pharmacological effects and lead to adverse events that have the potential to cause irreversible damage to the organism. Traditional methods to detect DDIs through biological or pharmacological analysis are time-consuming and expensive, therefore, there is an urgent need to develop computational methods to effectively predict drug-drug interactions. Currently, deep learning and knowledge graph techniques which can effectively extract features of entities have been widely utilized to develop DDI prediction methods. In this research, we aim to systematically review DDI prediction researches applying deep learning and graph knowledge. The available biomedical data and public databases related to drugs are firstly summarized in this review. Then, we discuss the existing drug-drug interactions prediction methods which have utilized deep learning and knowledge graph techniques and group them into three main classes: deep learning-based methods, knowledge graph-based methods, and methods that combine deep learning with knowledge graph. We comprehensively analyze the commonly used drug related data and various DDI prediction methods, and compare these prediction methods on benchmark datasets. Finally, we briefly discuss the challenges related to drug-drug interactions prediction, including asymmetric DDIs prediction and high-order DDI prediction.

Prediction of drug-disease associations based on reinforcement symmetric metric learning and graph convolution network.

Accurately identifying novel indications for drugs is crucial in drug research and discovery. Traditional drug discovery is costly and time-consuming. Computational drug repositioning can provide an effective strategy for discovering potential drug-disease associations. However, the known experimentally verified drug-disease associations is relatively sparse, which may affect the prediction performance of the computational drug repositioning methods. Moreover, while the existing drug-disease prediction method based on metric learning algorithm has achieved better performance, it simply learns features of drugs and diseases only from the drug-centered perspective, and cannot comprehensively model the latent features of drugs and diseases. In this study, we propose a novel drug repositioning method named RSML-GCN, which applies graph convolutional network and reinforcement symmetric metric learning to predict potential drug-disease associations. RSML-GCN first constructs a drug-disease heterogeneous network by integrating the association and feature information of drugs and diseases. Then, the graph convolutional network (GCN) is applied to complement the drug-disease association information. Finally, reinforcement symmetric metric learning with adaptive margin is designed to learn the latent vector representation of drugs and diseases. Based on the learned latent vector representation, the novel drug-disease associations can be identified by the metric function. Comprehensive experiments on benchmark datasets demonstrated the superior prediction performance of RSML-GCN for drug repositioning.

SPOP inhibits HBV transcription and replication by ubiquitination and degradation of HNF1α.

Hepatitis B virus (HBV) infection remains a significant public health burden worldwide. The persistence of covalently closed circular DNA (cccDNA) within the nucleus of infected hepatocytes is responsible for the failure of antiviral treatments. The ubiquitin proteasome system (UPS) has emerged as a promising antiviral target, as it can regulate HBV replication by promoting critical protein degradation in steps of viral life cycle. Speckle-type POZ protein (SPOP) is a critical adaptor for Cul3-RBX1 E3 ubiquitin ligase complex, but the effect of SPOP on HBV replication is less known. Here, we identified SPOP as a novel host antiviral factor against HBV infection. SPOP overexpression significantly inhibited the transcriptional activity of HBV cccDNA without affecting cccDNA level in HBV-infected HepG2-NTCP and primary human hepatocyte cells. Mechanism studies showed that SPOP interacted with hepatocyte nuclear factor 1α (HNF1α), and induced HNF1α degradation through host UPS pathway. Moreover, the antiviral role of SPOP was also confirmed in vivo. Together, our findings reveal that SPOP is a novel host factor which inhibits HBV transcription and replication by ubiquitination and degradation of HNF1α, providing a potential therapeutic strategy for the treatment of HBV infection.

Building Porous Ni(Salen)-Based Catalysts from Waste Styrofoam via Autocatalytic Coupling Chemistry for Heterogeneous Oxidation with Molecular Oxygen.

The development of robust and industrially viable catalysts from plastic waste is of great significance, and the facile construction of high performance heterogeneous catalyst systems for phenol-quinone conversions remains a grand challenge. Herein, a feasible strategy is demonstrated to reclaim Styrofoam into hierarchically porous nickel-salen-loaded hypercrosslinked polystyrene (PS@Ni-salen) catalysts with high activities through an unusual autocatalytic coupling route. The salen is immobilized onto PS chain by Friedel-Crafts alkylation of benzyl chloride derivatives, and the generated hydrogen chloride coordinately promotes the simultaneous crosslinking and bridge formation between aromatic rings via a Scholl coupling route, leading to hierarchically porous networks. After the metallization with Ni, the resultant networks exhibit high catalytic activity for the oxidation of 2,3,6-trimethylphenol to 2,3,5-trimethyl-1,4-benzoquinone under mild conditions (303 K, 1 bar of O2 ). This catalyst also demonstrates attractive recycling performance without an obvious loss of catalytic efficiency over five consecutive cycles. This methodology might provide a potential sustainable alternative to construct environmentally benign and cost-effective catalysts for specific organic transformation.

KGANSynergy: knowledge graph attention network for drug synergy prediction.

Combination therapy is widely used to treat complex diseases, particularly in patients who respond poorly to monotherapy. For example, compared with the use of a single drug, drug combinations can reduce drug resistance and improve the efficacy of cancer treatment. Thus, it is vital for researchers and society to help develop effective combination therapies through clinical trials. However, high-throughput synergistic drug combination screening remains challenging and expensive in the large combinational space, where an array of compounds are used. To solve this problem, various computational approaches have been proposed to effectively identify drug combinations by utilizing drug-related biomedical information. In this study, considering the implications of various types of neighbor information of drug entities, we propose a novel end-to-end Knowledge Graph Attention Network to predict drug synergy (KGANSynergy), which utilizes neighbor information of known drugs/cell lines effectively. KGANSynergy uses knowledge graph (KG) hierarchical propagation to find multi-source neighbor nodes for drugs and cell lines. The knowledge graph attention network is designed to distinguish the importance of neighbors in a KG through a multi-attention mechanism and then aggregate the entity's neighbor node information to enrich the entity. Finally, the learned drug and cell line embeddings can be utilized to predict the synergy of drug combinations. Experiments demonstrated that our method outperformed several other competing methods, indicating that our method is effective in identifying drug combinations.

SLHSD: hybrid scaffolding method based on short and long reads.

In genome assembly, scaffolding can obtain more complete and continuous scaffolds. Current scaffolding methods usually adopt one type of read to construct a scaffold graph and then orient and order contigs. However, scaffolding with the strengths of two or more types of reads seems to be a better solution to some tricky problems. Combining the advantages of different types of data is significant for scaffolding. Here, a hybrid scaffolding method (SLHSD) is present that simultaneously leverages the precision of short reads and the length advantage of long reads. Building an optimal scaffold graph is an important foundation for getting scaffolds. SLHSD uses a new algorithm that combines long and short read alignment information to determine whether to add an edge and how to calculate the edge weight in a scaffold graph. In addition, SLHSD develops a strategy to ensure that edges with high confidence can be added to the graph with priority. Then, a linear programming model is used to detect and remove remaining false edges in the graph. We compared SLHSD with other scaffolding methods on five datasets. Experimental results show that SLHSD outperforms other methods. The open-source code of SLHSD is available at https://github.com/luojunwei/SLHSD.

Polystyrene microplastics enhance the microcystin-LR-induced gonadal damage and reproductive endocrine disruption in zebrafish.

The coexistence of eutrophication and plastic pollution in the aquatic environment is becoming a realistic water pollution problem worldwide. To investigate the microcystin-LR (MC-LR) bioavailability and the underlying reproductive interferences in the presence of polystyrene microplastic (PSMPs), zebrafish (Danio rerio) were exposed to individual MC-LR (0, 1, 5, and 25 µg/L) and combined MC-LR + PSMPs (100 µg/L) for 60 d. Our results showed that the existence of PSMPs increased the accumulation of MC-LR in zebrafish gonads compared to the MC-LR-only group. In the MC-LR-only exposure group, seminiferous epithelium deterioration and widened intercellular spaces were observed in the testis, and basal membrane disintegration and zona pellucida invagination were noticed in the ovary. Moreover, the existence of PSMPs exacerbated these injuries. The results of sex hormone levels showed that PSMPs enhanced MC-LR-induced reproductive toxicity, which is tightly related to the abnormal increase of 17ß-estradiol (E2) and testosterone (T) levels. The changes of gnrh2, gnrh3, cyp19a1b, cyp11a, and lhr mRNA levels in the HPG axis further proved that MC-LR combined with PSMPs aggravated reproductive dysfunction. Our results revealed that PSMPs could increase the MC-LR bioaccumulation by serving as a carrier and exaggerate the MC-LR-induced gonadal damage and reproductive endocrine disruption in zebrafish.

Sub-chronic ammonia exposure induces hepatopancreatic damage, oxidative stress, and immune dysfunction in red swamp crayfish (Procambarus clarkii).

Ammonia, as one of the primary water pollutants in aquaculture, has been shown to induce a wide range of ecotoxicological effects on aquatic animals. In order to investigate the antioxidant and innate immune responses in crustaceans disrupted by ammonia, red swamp crayfish (Procambarus clarkii) were exposed to 0, 15, 30, and 50 mg/L total ammonia nitrogen for 30 d, the alterations of antioxidant responses as well as innate immunity were studied. The results showed that the severity of hepatopancreatic injury were aggravated by the increasing ammonia levels, which were mainly characterized by tubule lumen dilatation and vacuolization. The swollen mitochondria and disappeared mitochondria ridges suggested that oxidative stress induced by ammonia targets the mitochondria. Concurrently, enhanced MDA levels, and decreased GSH levels as well as the decreased transcription and activity of antioxidant enzymes, including SOD, CAT, and GPx were noticed, which suggested that high concentrations of ammonia exposure induce oxidative stress in P. clarkii. Furthermore, a significant decrease of the hemolymph ACP, AKP, and PO along with the significant downregulation of immune-related genes (ppo, hsp70, hsp90, alf1, ctl) jointly indicated that ammonia stress inhibited the innate immune function. Our findings demonstrated that sub-chronic ammonia stress induced hepatopancreatic injury and exert suppressive effects on the antioxidant capacity as well as innate immunity of P. clarkii. Our results provide a fundamental basis for the deleterious effects of ammonia stress on aquatic crustaceans.

Root-secreted (-)-loliolide modulates both belowground defense and aboveground flowering in Arabidopsis and tobacco.

Plant defense, growth, and reproduction can be modulated by chemicals emitted from neighboring plants, mainly via volatile aboveground signals. However, belowground signals and their underlying control mechanisms are largely unknown. Here, we experimentally demonstrate that the root-secreted carotenoid (-)-loliolide mediates both defensive and reproductive responses in wild-type Arabidopsis, a carotenoid-deficient Arabidopsis mutant (szl1-1), and tobacco (Nicotiana benthamiana). Wild-type Arabidopsis plants flower later than szl1-1, and they secrete (-)-loliolide into the soil, whereas szl1-1 roots do not. When Arabidopsis and tobacco occur together, wild-type Arabidopsis induces nicotine production and defense-related gene expression in tobacco, whereas szl1-1 impairs this induction but accelerates tobacco flowering. Furthermore, nicotine production and the expression of the key genes involved in nicotine biosynthesis (QPT, PMT1), plant defense (CAT1, SOD1, PR-2a, PI-II, TPI), and flowering (AP1, LFY, SOC1, FT3, FLC) are differently regulated by incubation with wild-type Arabidopsis and szl1-1 root exudates or (-)-loliolide. In particular, (-)-loliolide up-regulated flowering suppressors (FT3 and FLC) and transiently down-regulated flowering stimulators (AP1 and SOC1), delaying tobacco flowering. Therefore, root-secreted (-)-loliolide modulates plant belowground defense and aboveground flowering, yielding critical insights into plant-plant signaling interactions.

Deep learning approach for cancer subtype classification using high-dimensional gene expression data.

MOTIVATION: Studies have shown that classifying cancer subtypes can provide valuable information for a range of cancer research, from aetiology and tumour biology to prognosis and personalized treatment. Current methods usually adopt gene expression data to perform cancer subtype classification. However, cancer samples are scarce, and the high-dimensional features of their gene expression data are too sparse to allow most methods to achieve desirable classification results. RESULTS: In this paper, we propose a deep learning approach by combining a convolutional neural network (CNN) and bidirectional gated recurrent unit (BiGRU): our approach, DCGN, aims to achieve nonlinear dimensionality reduction and learn features to eliminate irrelevant factors in gene expression data. Specifically, DCGN first uses the synthetic minority oversampling technique algorithm to equalize data. The CNN can handle high-dimensional data without stress and extract important local features, and the BiGRU can analyse deep features and retain their important information; the DCGN captures key features by combining both neural networks to overcome the challenges of small sample sizes and sparse, high-dimensional features. In the experiments, we compared the DCGN to seven other cancer subtype classification methods using breast and bladder cancer gene expression datasets. The experimental results show that the DCGN performs better than the other seven methods and can provide more satisfactory classification results.

Particulate matter of air pollution may increase risk of kidney failure in IgA nephropathy.

IgA nephropathy (IgAN) is characterized by deposition of galactose-deficient IgA1 (Gd-IgA1) in glomerular mesangium associated with mucosal immune disorders. Since environmental pollution has been associated with the progression of chronic kidney disease in the general population, we specifically investigated the influence of exposure to fine particulate matter less than 2.5 µm in diameter (PM2.5) on IgAN progression. Patients with biopsy-proven primary IgAN were recruited from seven Chinese kidney centers. PM2.5 exposure from 1998 to 2016 was derived from satellite aerosol optical depth data and a total of 1,979 patients with IgAN, including 994 males were enrolled. The PM2.5 exposure levels for patients from different provinces varied but, in general, the PM2.5 exposure levels among patients from the north were higher than those among patients from the south. The severity of PM2.5 exposure in different regions was correlated with regional kidney failure burden. In addition, each 10 µg/m3 increase in annual average concentration of PM2.5 exposure before study entry (Hazard Ratio, 1.14; 95% confidence interval, 1.06-1.22) or time-varying PM2.5 exposure after study entry (1.10; 1.01-1.18) were associated with increased kidney failure risk after adjustment for age, gender, estimated glomerular filtration rate, urine protein, uric acid, hemoglobin, mean arterial pressure, Oxford classification, glucocorticoid and renin-angiotensin system blocker therapy. The associations were robust when the time period, risk factors of cardiovascular diseases or city size were further adjusted on the basis of the above model. Thus, our results suggest that PM2.5 is an independent risk factor for kidney failure in patients with IgAN, but these findings will require validation in more diverse populations and other geographic regions.

Distinct vulnerability to oxidative stress determines the ammonia sensitivity of crayfish (Procambarus clarkii) at different developmental stages.

Red swamp crayfish (Procambarus clarkii) has increasingly become a high-value freshwater product in China. During the intensive cultivation, excessive ammonia exposure is an important lethal factor of crayfish. We investigated the toxic effects and mechanisms of ammonia on crayfish at two different developmental stages. A preliminary ammonia stress test showed a 96-h LC50 of 135.10 mg/L and 299.61 mg/L for Stage_1 crayfish (8.47 ± 1.68 g) and Stage_2 crayfish (18.33 ± 2.41 g), respectively. During a prolonged ammonia exposure (up to 96 h), serum acid phosphatase and alkaline phosphatase showed a time-dependent decrease. Histological assessment indicated the degree of hepatopancreatic injury, which was mainly characterized as tubule lumen dilatation, degenerated tubule, vacuolization and dissolved hepatic epithelial cell, increased with exposure time. Enhanced malondialdehyde level and reduced antioxidant capacity of hepatopancreas were also observed. The mRNA expression and activity of catalase and superoxide dismutase showed an initial up-regulation within 24 h, and then gradually down-regulated with the exposure time. In the post-treatment recovery period, the Stage_2 crayfish exerted a stronger antioxidant and detoxification capacity than that of the Stage_1 crayfish, and thus quickly recovered from the ammonia exposure. Our findings provide a further understanding of the adverse effects of ammonia stress and suggest guidelines for water quality management during crayfish farming.

DACPGTN: Drug ATC Code Prediction Method Based on Graph Transformer Network for Drug Discovery.

The Anatomical Therapeutic Chemical (ATC) classification system is a drug classification scheme proposed by the World Health Organization, which is widely used for drug screening, repositioning, and similarity research. The ATC system assigns different ATC codes to drugs based on their anatomy, pharmacological, therapeutics and chemical properties. Predicting the ATC code of a given drug helps to understand the indication and potential toxicity of the drug, thus promoting its use in the therapeutic phase and accelerating its development. In this article, we propose an end-to-end model DACPGTN to predict the ATC code for the given drug. DACPGTN constructs composite features of drugs, diseases and targets by applying diverse biomedical information. Inspired by the application of Graph Transformer Network, we learn potential novel interactions among drugs diseases and targets from the known interactions to construct drug-target-disease heterogeneous networks containing comprehensive interaction information. Based on the constructed composite features and learned heterogeneous networks, we employ graph convolution network to generate the embedding of drug nodes, which are further used for the multi-label learning tasks in drug discovery. Experiments on the benchmark datasets demonstrate that the proposed DACPGTN model can achieve better prediction performance than the existing methods. The source codes of our method are available at https://github.com/Szhgege/DACPGTN.

Controlling the elasticity of polyacrylonitrile fibers via ionic liquids containing cyano-based anions.

As the predominant precursor for high-performance carbon fiber manufacturing, the fabrication of polyacrylonitrile (PAN)-based composite fibers attracts great interest. Ionic liquids (ILs) have recently been investigated for melt-spinning of ultrafine PAN fibers. The plasticizing properties of ILs are significantly affected by the structure of ILs and can be influenced by electronegativity, steric effects, etc. Herein, we report a facile strategy to control the elasticity of the PAN/ILs fibers by tuning the anion structure of ILs. Particularly, the ILs containing nitrile-rich groups exhibited enhanced plasticizing effect and nucleating ability on dissolving PAN components, achieving highly stretchable PAN/ILs fibers.

MultiGATAE: A Novel Cancer Subtype Identification Method Based on Multi-Omics and Attention Mechanism.

Cancer is one of the leading causes of death worldwide, which brings an urgent need for its effective treatment. However, cancer is highly heterogeneous, meaning that one cancer can be divided into several subtypes with distinct pathogenesis and outcomes. This is considered as the main problem which limits the precision treatment of cancer. Thus, cancer subtypes identification is of great importance for cancer diagnosis and treatment. In this work, we propose a deep learning method which is based on multi-omics and attention mechanism to effectively identify cancer subtypes. We first used similarity network fusion to integrate multi-omics data to construct a similarity graph. Then, the similarity graph and the feature matrix of the patient are input into a graph autoencoder composed of a graph attention network and omics-level attention mechanism to learn embedding representation. The K-means clustering method is applied to the embedding representation to identify cancer subtypes. The experiment on eight TCGA datasets confirmed that our proposed method performs better for cancer subtypes identification when compared with the other state-of-the-art methods. The source codes of our method are available at https://github.com/kataomoi7/multiGATAE.

A novel liver cancer diagnosis method based on patient similarity network and DenseGCN.

Liver cancer is the main malignancy in terms of mortality rate, accurate diagnosis can help the treatment outcome of liver cancer. Patient similarity network is an important information which helps in cancer diagnosis. However, recent works rarely take patient similarity into consideration. To address this issue, we constructed patient similarity network using three liver cancer omics data, and proposed a novel liver cancer diagnosis method consisted of similarity network fusion, denoising autoencoder and dense graph convolutional neural network to capitalize on patient similarity network and multi omics data. We compared our proposed method with other state-of-the-art methods and machine learning methods on TCGA-LIHC dataset to evaluate its performance. The results confirmed that our proposed method surpasses these comparison methods in terms of all the metrics. Especially, our proposed method has attained an accuracy up to 0.9857.

Flux upcycling of spent NMC 111 to nickel-rich NMC cathodes in reciprocal ternary molten salts.

The proper handling of end-of-life (EOL) lithium-ion batteries (LIBs) has become an urgent and challenging issue with the surging use of LIBs, in which recovering high-value cathodes not only relieves the pressure on the raw material supply chain but also minimizes environmental pollution. Beyond direct recycling of spent cathodes to their pristine states, the direct upcycling of spent cathodes to the next-generation cathodes is of great significance to maximize the value of spent materials and to sustain the fast development of LIBs. Herein, a "reciprocal ternary molten salts" (RTMS) system was developed to directly upcycle spent NMC 111 to Ni-rich NMCs by simultaneously realizing the addition of Ni and the relithiation of Li in spent NMC 111. After RTMS flux upcycling, the obtained Ni-rich NMCs exhibited an α-NaFeO2-type layered structure, restored Li content, and excellent performance, which is very similar to that of the pristine NMC 622.

New-Generation Carbon-Capture Ionic Liquids Regulated by Metal-Ion Coordination.

Development of efficient carbon capture-and-release technologies with minimal energy input is a long-term challenge in mitigating CO2 emissions, especially via CO2 chemisorption driven by engineered chemical bond construction. Herein, taking advantage of the structural diversity of ionic liquids (ILs) in tuning their physical and chemical properties, precise reaction energy regulation of CO2 chemisorption was demonstrated deploying metal-ion-amino-based ionic liquids (MAILs) as absorbents. The coordination ability of different metal sites (Cu, Zn, Co, Ni, and Mg) to amines was harnessed to achieve fine-tuning on stability constants of the metal ion-amine complexes, acting as the corresponding cations in the construction of diverse ILs coupled with CO2 -philic anions. The as-afforded MAILs exhibited efficient and controllable CO2 release behavior with great reduction in energy input and minimal sacrifice on CO2 uptake capacity. This coordination-regulated approach offers new prospects for the development of ILs-based systems and beyond towards energy-efficient carbon capture technologies.

BreakNet: detecting deletions using long reads and a deep learning approach.

BACKGROUND: Structural variations (SVs) occupy a prominent position in human genetic diversity, and deletions form an important type of SV that has been suggested to be associated with genetic diseases. Although various deletion calling methods based on long reads have been proposed, a new approach is still needed to mine features in long-read alignment information. Recently, deep learning has attracted much attention in genome analysis, and it is a promising technique for calling SVs. RESULTS: In this paper, we propose BreakNet, a deep learning method that detects deletions by using long reads. BreakNet first extracts feature matrices from long-read alignments. Second, it uses a time-distributed convolutional neural network (CNN) to integrate and map the feature matrices to feature vectors. Third, BreakNet employs a bidirectional long short-term memory (BLSTM) model to analyse the produced set of continuous feature vectors in both the forward and backward directions. Finally, a classification module determines whether a region refers to a deletion. On real long-read sequencing datasets, we demonstrate that BreakNet outperforms Sniffles, SVIM and cuteSV in terms of their F1 scores. The source code for the proposed method is available from GitHub at https://github.com/luojunwei/BreakNet . CONCLUSIONS: Our work shows that deep learning can be combined with long reads to call deletions more effectively than existing methods.

A Validation Study Comparing Risk Prediction Models of IgA Nephropathy.

We aimed to validate three IgAN risk models proposed by an international collaborative study and another CKD risk model generated by an extended CKD cohort with our multicenter Chinese IgAN cohort. Biopsy-proven IgAN patients with an eGFR ≥15 ml/min/1.73 m2 at baseline and a minimum follow-up of 6 months were enrolled. The primary outcomes were a composite outcome (50% decline in eGFR or ESRD) and ESRD. The performance of those models was assessed using discrimination, calibration, and reclassification. A total of 2,300 eligible cases were enrolled. Of them, 288 (12.5%) patients reached composite outcome and 214 (9.3%) patients reached ESRD during a median follow-up period of 30 months. Using the composite outcome for analysis, the Clinical, Limited, Full, and CKD models had relatively good performance with similar C statistics (0.81, 0.81, 0.82, and 0.82, respectively). While using ESRD as the end point, the four prediction models had better performance (all C statistics > 0.9). Furthermore, subgroup analysis showed that the models containing clinical and pathological variables (Full model and Limited model) had better discriminatory abilities than the models including only clinical indicators (Clinical model and CKD model) in low-risk patients characterized by higher baseline eGFR (≥60 ml/min/1.73 m2). In conclusion, we validated recently reported IgAN and CKD risk models in our Chinese IgAN cohort. Compared to pure clinical models, adding pathological variables will increase performance in predicting ESRD in low-risk IgAN patients with baseline eGFR ≥60 ml/min/1.73 m2.