Graph representation learning in bioinformatics: trends, methods and applications.

Graph is a natural data structure for describing complex systems, which contains a set of objects and relationships. Ubiquitous real-life biomedical problems can be modeled as graph analytics tasks. Machine learning, especially deep learning, succeeds in vast bioinformatics scenarios with data represented in Euclidean domain. However, rich relational information between biological elements is retained in the non-Euclidean biomedical graphs, which is not learning friendly to classic machine learning methods. Graph representation learning aims to embed graph into a low-dimensional space while preserving graph topology and node properties. It bridges biomedical graphs and modern machine learning methods and has recently raised widespread interest in both machine learning and bioinformatics communities. In this work, we summarize the advances of graph representation learning and its representative applications in bioinformatics. To provide a comprehensive and structured analysis and perspective, we first categorize and analyze both graph embedding methods (homogeneous graph embedding, heterogeneous graph embedding, attribute graph embedding) and graph neural networks. Furthermore, we summarize their representative applications from molecular level to genomics, pharmaceutical and healthcare systems level. Moreover, we provide open resource platforms and libraries for implementing these graph representation learning methods and discuss the challenges and opportunities of graph representation learning in bioinformatics. This work provides a comprehensive survey of emerging graph representation learning algorithms and their applications in bioinformatics. It is anticipated that it could bring valuable insights for researchers to contribute their knowledge to graph representation learning and future-oriented bioinformatics studies.

Accurate prediction of drug combination risk levels based on relational graph convolutional network and multi-head attention.

BACKGROUND: Accurately identifying the risk level of drug combinations is of great significance in investigating the mechanisms of combination medication and adverse reactions. Most existing methods can only predict whether there is an interaction between two drugs, but cannot directly determine their accurate risk level. METHODS: In this study, we propose a multi-class drug combination risk prediction model named AERGCN-DDI, utilizing a relational graph convolutional network with a multi-head attention mechanism. Drug-drug interaction events with varying risk levels are modeled as a heterogeneous information graph. Attribute features of drug nodes and links are learned based on compound chemical structure information. Finally, the AERGCN-DDI model is proposed to predict drug combination risk level based on heterogenous graph neural network and multi-head attention modules. RESULTS: To evaluate the effectiveness of the proposed method, five-fold cross-validation and ablation study were conducted. Furthermore, we compared its predictive performance with baseline models and other state-of-the-art methods on two benchmark datasets. Empirical studies demonstrated the superior performances of AERGCN-DDI. CONCLUSIONS: AERGCN-DDI emerges as a valuable tool for predicting the risk levels of drug combinations, thereby aiding in clinical medication decision-making, mitigating severe drug side effects, and enhancing patient clinical prognosis.

miR-541 is associated with the prognosis of liver cirrhosis and directly targets JAG2 to inhibit the activation of hepatic stellate cells.

BACKGROUND: The activation of hepatic stellate cells (HSCs) has been emphasized as a leading event of the pathogenesis of liver cirrhosis, while the exact mechanism of its activation is largely unknown. Furthermore, the novel non-invasive predictors of prognosis in cirrhotic patients warrant more exploration. miR-541 has been identified as a tumor suppressor in hepatocellular carcinoma and a regulator of fibrotic disease, such as lung fibrosis and renal fibrosis. However, its role in liver cirrhosis has not been reported. METHODS: Real-time PCR was used to detect miR-541 expression in the liver tissues and sera of liver cirrhosis patients and in the human LX-2. Gain- and loss-of-function assays were performed to evaluate the effects of miR-541 on the activation of LX-2. Bioinformatics analysis and a luciferase reporter assay were conducted to investigate the target gene of miR-541. RESULTS: miR-541 was downregulated in the tissues and sera of patients with liver cirrhosis, which was exacerbated by deteriorating disease severity. Importantly, the lower expression of miR-541 was associated with more episodes of complications including ascites and hepatic encephalopathy, a shorter overall lifespan, and decompensation-free survival. Moreover, multivariate Cox's regression analysis verified lower serum miR-541 as an independent risk factor for liver-related death in cirrhotic patients (HR = 0.394; 95% CI: 0.164-0.947; P = 0.037). miR-541 was also decreased in LX-2 cells activated by TGF-ß and the overexpression of miR-541 inhibited the proliferation, activation and hydroxyproline secretion of LX-2 cells. JAG2 is an important ligand of Notch signaling and was identified as a direct target gene of miR-541. The expression of JAG2 was upregulated in the liver tissues of cirrhotic patients and was inversely correlated with miR-541 levels. A rescue assay further confirmed that JAG2 was involved in the function of miR-541 when regulating LX-2 activation and Notch signaling. CONCLUSIONS: Dysregulation of miR-541/JAG2 axis might be a as a new mechanism of liver fibrosis, and miR-541 could serve as a novel non-invasive biomarker and therapeutic targets for liver cirrhosis.

Analysis of Postoperative Urinary Incontinence and Influencing Factors of Transurethral Holmium Laser Enucleation of the Prostate.

INTRODUCTION: Aims of the study were to investigate the related factors of urinary incontinence after transurethral holmium laser enucleation of the prostate (HoLEP) and to provide guidance for clinical urinary control of HoLEP. METHODS: The clinical data of 548 patients who underwent HoLEP were retrospectively analyzed. The patients were followed up for the occurrence of urinary incontinence in the short term (2 weeks), medium term (3 months), and long term (6 months) after HoLEP. RESULTS: Among the 548 benign prostatic hyperplasia patients, 79 cases (14.42%) had urinary incontinence at 2 weeks, 19 cases (3.47%) at 3 months, and 2 cases (0.36%) at 6 months after surgery. Logistic regression analysis showed that age, prostate volume, diabetes mellitus, operation time, prostate tissue weight, and histological prostatitis were risk factors for recent urinary incontinence (p < 0.05). Age, diabetes, and operation time were risk factors for mid-term urinary incontinence (p < 0.05). The incidence of long-term urinary incontinence was low and no risk factor analysis was performed. CONCLUSIONS: For good urinary control after HoLEP, in addition to surgery-related factors such as surgical skills, proficiency, and precise anatomy, patients' risk factors should also be paid attention to in order to improve postoperative urinary control more effectively and reduce the incidence of urinary incontinence.

Poor pretransplantation minimal residual disease clearance as an independent prognostic risk factor for survival in myelodysplastic syndrome with excess blasts: A multicenter, retrospective cohort study.

BACKGROUND: Minimal residual disease (MRD) is an important prognostic factor for survival in adults with acute leukemia. The role of pretransplantation MRD status in myelodysplastic syndrome with excess blasts (MDS-EB) is unknown. This study retrospectively analyzed the relationship between pretransplantation MRD status and long-term survival. MATERIALS AND METHODS: Patients with MDS-EB who underwent allogeneic hematopoietic stem cell transplantation (allo-HSCT) from March 5, 2005, to November 8, 2020, were included. The relationship between pretransplantation MRD status and long-term survival was analyzed using univariate and multivariate logistic regression models. RESULTS: Of 220 patients with MDS-EB who underwent allo-HSCT, 198 were eligible for inclusion in this multicenter, retrospective cohort study. Complete remission was attained in 121 (61.1%) patients, and 103 patients underwent detection of MRD pretransplantation, with 67 patients being MRD-positive and 36 patients being MRD-negative. The median follow-up time was 16 months, the median age was 41 years (6-65 years), and 58% of the patients were men. The 3-year disease-free survival (DFS) and overall survival (OS) probabilities for all patients were 70.1% and 72.9%, respectively. For patients in complete remission, the 3-year DFS and OS probabilities were 72.2% and 74.8%, respectively. Further analysis found that the 3-year DFS rates of MRD-negative and MRD-positive patients were 85.6% and 66.5% (p = .045), respectively, whereas the 3-year OS rates were 91.3% and 66.4% (p = .035), respectively. Univariate and multivariate analyses showed that poor pretransplantation MRD clearance was an independent prognostic risk factor for DFS and OS. CONCLUSION: Poor pretransplantation MRD clearance is an independent prognostic risk factor for long-term survival after allo-HSCT for patients with MDS-EB. PLAIN LANGUAGE SUMMARY: Poor minimal residual disease clearance pretransplanation is an independent prognostic risk factor for long-term survival after allogeneic hematopoietic stem cell transplantation for patients with myelodysplastic syndrome with excess blasts.

MeSHHeading2vec: a new method for representing MeSH headings as vectors based on graph embedding algorithm.

Effectively representing Medical Subject Headings (MeSH) headings (terms) such as disease and drug as discriminative vectors could greatly improve the performance of downstream computational prediction models. However, these terms are often abstract and difficult to quantify. In this paper, we converted the MeSH tree structure into a relationship network and applied several graph embedding algorithms on it to represent these terms. Specifically, the relationship network consisting of nodes (MeSH headings) and edges (relationships), which can be constructed by the tree num. Then, five graph embedding algorithms including DeepWalk, LINE, SDNE, LAP and HOPE were implemented on the relationship network to represent MeSH headings as vectors. In order to evaluate the performance of the proposed methods, we carried out the node classification and relationship prediction tasks. The results show that the MeSH headings characterized by graph embedding algorithms can not only be treated as an independent carrier for representation, but also can be utilized as additional information to enhance the representation ability of vectors. Thus, it can serve as an input and continue to play a significant role in any computational models related to disease, drug, microbe, etc. Besides, our method holds great hope to inspire relevant researchers to study the representation of terms in this network perspective.

Impact of a novel prognostic model on allogeneic hematopoietic stem cell transplantation outcomes in patients with CMML.

Chronic myelomonocytic leukemia (CMML) is a clonal hematopoietic stem cell malignancy, and allogeneic hematopoietic stem cell transplantation (allo-HSCT) is the only curable treatment. The outcomes after transplant are influenced by both disease characteristics and patient comorbidities. To develop a novel prognostic model to predict the post-transplant survival of CMML patients, we identified risk factors by applying univariable and multivariable Cox proportional hazards regression to a derivation cohort. In multivariable analysis, advanced age (hazard ratio [HR] 3.583), leukocyte count (HR 3.499), anemia (HR 3.439), bone marrow blast cell count (HR 2.095), and no chronic graft versus host disease (cGVHD; HR 4.799) were independently associated with worse survival. A novel prognostic model termed ABLAG (Age, Blast, Leukocyte, Anemia, cGVHD) was developed and the points were assigned according to the regression equation. The patients were categorized into low risk (0-1), intermediate risk (2, 3), and high risk (4-6) three groups and the 3-year overall survival (OS) were 93.3% (95%CI, 61%-99%), 78.9% (95%CI, 60%-90%), and 51.6% (95%CI, 32%-68%; p < .001), respectively. In internal and external validation cohort, the area under the receiver operating characteristic (ROC) curves of the ABLAG model were 0.829 (95% CI, 0.776-0.902) and 0.749 (95% CI, 0.684-0.854). Compared with existing models designed for the nontransplant setting, calibration plots, and decision curve analysis showed that the ABLAG model revealed a high consistency between predicted and observed outcomes and patients could benefit from this model. In conclusion, combining disease and patient characteristic, the ABLAG model provides better survival stratification for CMML patients receiving allo-HSCT.

Type I Photosensitizer Targeting G-Quadruplex RNA Elicits Augmented Immunity for Cancer Ablation.

Type I photodynamic therapy (PDT) represents a promising treatment modality for tumors with intrinsic hypoxia. However, type I photosensitizers (PSs), especially ones with near infrared (NIR) absorption, are limited and their efficacy needs improvement via new targeting tactics. We develop a NIR type I PS by engineering acridinium derived donor-π-acceptor systems. The PS exhibits an exclusive type I PDT mechanism due to effective intersystem crossing and disfavored energy transfer to O2 , and shows selective binding to G-quadruplexes (G4s) via hydrogen bonds identified by a molecular docking study. Moreover, it enables fluorogenic detection of G4s and efficient O2 ⋅- production in hypoxic conditions, leading to immunogenic cell death and substantial variations of gene expression in RNA sequencing. Our strategy demonstrates augmented antitumor immunity for effective ablation of immunogenic cold tumor, highlighting its potential of RNA-targeted type I PDT in precision cancer therapy.

DeepWalk based method to predict lncRNA-miRNA associations via lncRNA-miRNA-disease-protein-drug graph.

BACKGROUND: Long non-coding RNAs (lncRNAs) play a crucial role in diverse biological processes and have been confirmed to be concerned with various diseases. Largely uncharacterized of the physiological role and functions of lncRNA remains. MicroRNAs (miRNAs), which are usually 20-24 nucleotides, have several critical regulatory parts in cells. LncRNA can be regarded as a sponge to adsorb miRNA and indirectly regulate transcription and translation. Thus, the identification of lncRNA-miRNA associations is essential and valuable. RESULTS: In our work, we present DWLMI to infer the potential associations between lncRNAs and miRNAs by representing them as vectors via a lncRNA-miRNA-disease-protein-drug graph. Specifically, DeepWalk can be used to learn the behavior representation of vertices. The methods of fingerprint, k-mer and MeSH descriptors were mainly used to learn the attribute representation of vertices. By combining the above two kinds of information, unknown lncRNA-miRNA associations can be predicted by the random forest classifier. Under the five-fold cross-validation, the proposed DWLMI model obtained an average prediction accuracy of 95.22% with a sensitivity of 94.35% at the AUC of 98.56%. CONCLUSIONS: The experimental results demonstrated that DWLMI can effectively predict the potential lncRNA-miRNA associated pairs, and the results can provide a new insight for related non-coding RNA researchers in the field of combing biology big data with deep learning.

A learning-based method to predict LncRNA-disease associations by combining CNN and ELM.

BACKGROUND: lncRNAs play a critical role in numerous biological processes and life activities, especially diseases. Considering that traditional wet experiments for identifying uncovered lncRNA-disease associations is limited in terms of time consumption and labor cost. It is imperative to construct reliable and efficient computational models as addition for practice. Deep learning technologies have been proved to make impressive contributions in many areas, but the feasibility of it in bioinformatics has not been adequately verified. RESULTS: In this paper, a machine learning-based model called LDACE was proposed to predict potential lncRNA-disease associations by combining Extreme Learning Machine (ELM) and Convolutional Neural Network (CNN). Specifically, the representation vectors are constructed by integrating multiple types of biology information including functional similarity and semantic similarity. Then, CNN is applied to mine both local and global features. Finally, ELM is chosen to carry out the prediction task to detect the potential lncRNA-disease associations. The proposed method achieved remarkable Area Under Receiver Operating Characteristic Curve of 0.9086 in Leave-one-out cross-validation and 0.8994 in fivefold cross-validation, respectively. In addition, 2 kinds of case studies based on lung cancer and endometrial cancer indicate the robustness and efficiency of LDACE even in a real environment. CONCLUSIONS: Substantial results demonstrated that the proposed model is expected to be an auxiliary tool to guide and assist biomedical research, and the close integration of deep learning and biology big data will provide life sciences with novel insights.

GBDR: a Bayesian model for precise prediction of pathogenic microorganisms using 16S rRNA gene sequences.

BACKGROUND: Recent evidences have suggested that human microorganisms participate in important biological activities in the human body. The dysfunction of host-microbiota interactions could lead to complex human disorders. The knowledge on host-microbiota interactions can provide valuable insights into understanding the pathological mechanism of diseases. However, it is time-consuming and costly to identify the disorder-specific microbes from the biological "haystack" merely by routine wet-lab experiments. With the developments in next-generation sequencing and omics-based trials, it is imperative to develop computational prediction models for predicting microbe-disease associations on a large scale. RESULTS: Based on the known microbe-disease associations derived from the Human Microbe-Disease Association Database (HMDAD), the proposed model shows reliable performance with high values of the area under ROC curve (AUC) of 0.9456 and 0.8866 in leave-one-out cross validations and five-fold cross validations, respectively. In case studies of colorectal carcinoma, 80% out of the top-20 predicted microbes have been experimentally confirmed via published literatures. CONCLUSION: Based on the assumption that functionally similar microbes tend to share the similar interaction patterns with human diseases, we here propose a group based computational model of Bayesian disease-oriented ranking to prioritize the most potential microbes associating with various human diseases. Based on the sequence information of genes, two computational approaches (BLAST+ and MEGA 7) are leveraged to measure the microbe-microbe similarity from different perspectives. The disease-disease similarity is calculated by capturing the hierarchy information from the Medical Subject Headings (MeSH) data. The experimental results illustrate the accuracy and effectiveness of the proposed model. This work is expected to facilitate the characterization and identification of promising microbial biomarkers.

Jiella sonneratiae sp. nov., a novel endophytic bacterium isolated from bark of Sonneratia apetala.

A new endophytic bacterium, designated strain MQZ13P-4T was isolated from Sonneratia apetala collected from Maowei sea Mangrove Nature Reserve in Guangxi Zhuang Autonomous Region, PR China. The 16S rRNA gene sequence similarity between strain MQZ13P-4T and its closest phylogenetic neighbour Jiella endophytica CBS5Q-3T was 97.9 %. Phylogenetic analyses using 16S rRNA gene sequences and whole-genome sequences showed that strain MQZ13P-4T formed a distinct lineage with Jiella endophytica CBS5Q-3T, Jiella pacifica 40Bstr34T and Jiella aquimaris JCM 30119T. The draft genome of strain MQZ13P-4T was 5 153 243 bp in size and its DNA G+C content was 68.1 mol%. Comparative genome analysis revealed that the average nucleotide identity, digital DNA-DNA hybridization and average amino acid identity values among strain MQZ13P-4T and other related species were below the cut-off levels of 95, 70 and 95.5 %, respectively. The cell-wall peptidoglycan of strain MQZ13P-4T contained meso-diaminopimelic acid as the diagnostic diamino acid. The respiratory quinone was Q-10. The major cellular fatty acid was C18 : 1 ω7c. The polar lipids comprised phosphatidylcholine, phosphatidylethanolamine, phosphatidylmonomethylethanolamine, phosphatidylglycerol, diphosphatidylglycerol, two unidentified aminolipids and two unidentified lipids. Strain MQZ13P-4T had a typical chemical compositions of fatty acids, lipids, quinones and diagnostic diamino acid for Jiella species, but could be distinguished from known species of the genus Jiella. Based on polyphasic evidence, strain MQZ13P-4T represents novel species of the genus Jiella, for which the name Jiella sonneratiae sp. nov. is proposed. The type strain is MQZ13P-4T (=CGMCC 1.18727T=JCM 34333T).

In silico drug repositioning using deep learning and comprehensive similarity measures.

BACKGROUND: Drug repositioning, meanings finding new uses for existing drugs, which can accelerate the processing of new drugs research and development. Various computational methods have been presented to predict novel drug-disease associations for drug repositioning based on similarity measures among drugs and diseases. However, there are some known associations between drugs and diseases that previous studies not utilized. METHODS: In this work, we develop a deep gated recurrent units model to predict potential drug-disease interactions using comprehensive similarity measures and Gaussian interaction profile kernel. More specifically, the similarity measure is used to exploit discriminative feature for drugs based on their chemical fingerprints. Meanwhile, the Gaussian interactions profile kernel is employed to obtain efficient feature of diseases based on known disease-disease associations. Then, a deep gated recurrent units model is developed to predict potential drug-disease interactions. RESULTS: The performance of the proposed model is evaluated on two benchmark datasets under tenfold cross-validation. And to further verify the predictive ability, case studies for predicting new potential indications of drugs were carried out. CONCLUSION: The experimental results proved the proposed model is a useful tool for predicting new indications for drugs or new treatments for diseases, and can accelerate drug repositioning and related drug research and discovery.

Serum Inflammatory Factor Profiles in the Pathogenesis of High-Altitude Polycythemia and Mechanisms of Acclimation to High Altitudes.

High-altitude polycythemia (HAPC) is a common aspect of chronic mountain sickness (CMS) caused by hypoxia and is the main cause of other symptoms associated with CMS. However, its pathogenesis and the mechanisms of high-altitude acclimation have not been fully elucidated. Exposure to high altitude is associated with elevated inflammatory mediators. In this study, the subjects were recruited and placed into a plain control (PC) group, plateau control (PUC) group, early HAPC (eHAPC) group, or a confirmed HAPC (cHAPC) group. Serum samples were collected, and inflammatory factors were measured by a novel antibody array methodology. The serum levels of interleukin-2 (IL-2), interleukin-3 (IL-3), and macrophage chemoattractant protein-1 (MCP-1) in the eHAPC group and the levels of interleukin-1 beta (IL-1 beta), IL-2, IL-3, tumor necrosis factor-alpha (TNF-alpha), MCP-1, and interleukin-16 (IL-16) in the cHAPC group were higher than those in the PUC group. More interestingly, the expression of IL-1 beta, IL-2, IL-3, TNF-alpha, MCP-1, and IL-16 in the PUC group showed a remarkable lower value than that in the PC group. These results suggest that these six factors might be involved in the pathogenesis of HAPC as well as acclimation to high altitudes. Altered inflammatory factors might be new biomarkers for HAPC and for high-altitude acclimation.

NEMPD: a network embedding-based method for predicting miRNA-disease associations by preserving behavior and attribute information.

BACKGROUND: As an important non-coding RNA, microRNA (miRNA) plays a significant role in a series of life processes and is closely associated with a variety of Human diseases. Hence, identification of potential miRNA-disease associations can make great contributions to the research and treatment of Human diseases. However, to our knowledge, many existing computational methods only utilize the single type of known association information between miRNAs and diseases to predict their potential associations, without focusing on their interactions or associations with other types of molecules. RESULTS: In this paper, we propose a network embedding-based method for predicting miRNA-disease associations by preserving behavior and attribute information. Firstly, a heterogeneous network is constructed by integrating known associations among miRNA, protein and disease, and the network representation method Learning Graph Representations with Global Structural Information (GraRep) is implemented to learn the behavior information of miRNAs and diseases in the network. Then, the behavior information of miRNAs and diseases is combined with the attribute information of them to represent miRNA-disease association pairs. Finally, the prediction model is established based on the Random Forest algorithm. Under the five-fold cross validation, the proposed NEMPD model obtained average 85.41% prediction accuracy with 80.96% sensitivity at the AUC of 91.58%. Furthermore, the performance of NEMPD is also validated by the case studies. Among the top 50 predicted disease-related miRNAs, 48 (breast neoplasms), 47 (colon neoplasms), 47 (lung neoplasms) were confirmed by two other databases. CONCLUSIONS: The proposed NEMPD model has a good performance in predicting the potential associations between miRNAs and diseases, and has great potency in the field of miRNA-disease association prediction in the future.

RPI-SE: a stacking ensemble learning framework for ncRNA-protein interactions prediction using sequence information.

BACKGROUND: The interactions between non-coding RNAs (ncRNA) and proteins play an essential role in many biological processes. Several high-throughput experimental methods have been applied to detect ncRNA-protein interactions. However, these methods are time-consuming and expensive. Accurate and efficient computational methods can assist and accelerate the study of ncRNA-protein interactions. RESULTS: In this work, we develop a stacking ensemble computational framework, RPI-SE, for effectively predicting ncRNA-protein interactions. More specifically, to fully exploit protein and RNA sequence feature, Position Weight Matrix combined with Legendre Moments is applied to obtain protein evolutionary information. Meanwhile, k-mer sparse matrix is employed to extract efficient feature of ncRNA sequences. Finally, an ensemble learning framework integrated different types of base classifier is developed to predict ncRNA-protein interactions using these discriminative features. The accuracy and robustness of RPI-SE was evaluated on three benchmark data sets under five-fold cross-validation and compared with other state-of-the-art methods. CONCLUSIONS: The results demonstrate that RPI-SE is competent for ncRNA-protein interactions prediction task with high accuracy and robustness. It's anticipated that this work can provide a computational prediction tool to advance ncRNA-protein interactions related biomedical research.

Incidence and risk factors for post-ERCP pancreatitis in pancreas divisum patients without chronic pancreatitis.

Aims: The studies on post-endoscopic retrograde cholangiopancreatography (ERCP) pancreatitis (PEP) in pancreas divisum (PD) patients without chronic pancreatitis (CP) are rare. In this study, we aimed to evaluate the incidence of PEP in PD patients without CP and the risk and protective factors for PEP.Methods: Consecutive patients with symptomatic PD that underwent ERCP from January 2005 to December 2017 were retrospectively analyzed. The patients were divided into PD without CP group and CP group. The basic information and medical records of patients were collected. The risk and protective factors for PEP in PD patients without CP were analyzed by univariate logistic analysis.Results: A total of 89 ERCP procedures were performed in 51 PD patients without CP, and 249 procedures in 136 patients with CP. The incidence of PEP was significantly higher in PD patients without CP than those with CP (15.7% vs. 5.6%, p = .005). Female gender were independent risk factors for PEP, while dorsal duct stent placement was a protective factor.Conclusion: CP may be a protective factor against PEP in PD patients. Female was a risk factor for PEP in PD patients and dorsal duct stent placement was a preventive factor that reduced the incidence of PEP in PD patients without CP.

Extracellular Vesicles with Possible Roles in Gut Intestinal Tract Homeostasis and IBD.

The intestinal tract consists of various types of cells, such as epithelial cells, Paneth cells, macrophages, and lymphocytes, which constitute the intestinal immune system and play a significant role in maintaining intestinal homeostasis by producing antimicrobial materials and controlling the host-commensal balance. Various studies have found that the dysfunction of intestinal homeostasis contributes to the pathogenesis of inflammatory bowel disease (IBD). As a novel mediator, extracellular vesicles (EVs) have been recognized as effective communicators, not only between cells but also between cells and the organism. In recent years, EVs have been regarded as vital characters for dysregulated homeostasis and IBD in either the etiology or the pathology of intestinal inflammation. Here, we review recent studies on EVs associated with intestinal homeostasis and IBD and discuss their source, cargo, and origin, as well as their therapeutic effects on IBD, which mainly include artificial nanoparticles and EVs derived from microorganisms.

A deep learning-based method for drug-target interaction prediction based on long short-term memory neural network.

BACKGROUND: The key to modern drug discovery is to find, identify and prepare drug molecular targets. However, due to the influence of throughput, precision and cost, traditional experimental methods are difficult to be widely used to infer these potential Drug-Target Interactions (DTIs). Therefore, it is urgent to develop effective computational methods to validate the interaction between drugs and target. METHODS: We developed a deep learning-based model for DTIs prediction. The proteins evolutionary features are extracted via Position Specific Scoring Matrix (PSSM) and Legendre Moment (LM) and associated with drugs molecular substructure fingerprints to form feature vectors of drug-target pairs. Then we utilized the Sparse Principal Component Analysis (SPCA) to compress the features of drugs and proteins into a uniform vector space. Lastly, the deep long short-term memory (DeepLSTM) was constructed for carrying out prediction. RESULTS: A significant improvement in DTIs prediction performance can be observed on experimental results, with AUC of 0.9951, 0.9705, 0.9951, 0.9206, respectively, on four classes important drug-target datasets. Further experiments preliminary proves that the proposed characterization scheme has great advantage on feature expression and recognition. We also have shown that the proposed method can work well with small dataset. CONCLUSION: The results demonstration that the proposed approach has a great advantage over state-of-the-art drug-target predictor. To the best of our knowledge, this study first tests the potential of deep learning method with memory and Turing completeness in DTIs prediction.

Identification of Fusarium ipomoeae as the causative agent of leaf spot disease in Bletilla striata in China.

Bletilla striata (Thunb.) Rchb. f. (Orchidaceae) is traditionally used for hemostasis and detumescence in China. In April 2019, a leaf spot disease on B. striata was observed in plant nurseries in Guilin, Guangxi Province, China, with an estimated incidence of ~30%. Initial symptoms include the appearance of circular or irregular brown spots on leaf surfaces, which progressively expand into large, dark brown, necrotic areas. As lesions coalesce, large areas of the leaf die, ultimately resulting in abscission. To isolate the pathogen, representative samples exhibiting symptoms were collected, leaf tissues (5 × 5 mm) were cut from the junction of diseased and healthy tissue, surface-disinfected in 1% sodium hypochlorite solution for 2 min, rinsed three times in sterile water, plated on potato dextrose agar (PDA) medium, and incubated at 28°C (12-h light-dark cycle) for 3 days. Hyphal tips from recently germinated spores were transferred to PDA to obtain pure cultures. Nine fungal isolates with similar morphological characteristics were obtained. Colonies on PDA were villose, had a dense growth of aerial mycelia and appeared pinkish white from above and greyish orange at the center and pinkish-white at the margin on the underside. Macroconidia were smooth, and hyaline, with a dorsiventral curvature, hooked to tapering apical cells, and 3- to 5-septate. Three-septate macroconidia were 21.2 to 32.1 × 2.4 to 3.9 µm (mean ± SD: 26.9 ± 2.5 × 3.2 ± 0.4 µm, n = 30); 4-septate macroconidia were 29.5 to 38.9 × 3.0 to 4.3 µm (mean ± SD: 33.5 ± 2.6 × 3.6 ± 0.3 µm, n = 40); and 5-septate macroconidia were 39.3 to 55.6 × 4.0 to 5.4 µm (mean ± SD: 48.0 ± 3.9 × 4.5 ± 0.3 µm, n = 50). These morphological characteristics were consistent with F. ipomoeae, a member of the Fusarium incarnatum-equiseti species complex (FIESC) (Wang et al. 2019). To confirm the fungal isolate's identification, the genomic DNA of the single-spore isolate BJ-22.3 was extracted using the CTAB method (Guo et al. 2000). The internal transcribed space (ITS) region of rDNA, translation elongation factor-1 alpha (TEF-1α), and partial RNA polymerase second largest subunit (RPB2) were amplified using primer pairs [ITS1/ITS4 (White et al. 1990), EF-1/EF-2 (O'Donnell et al. 1998), and 5f2/11ar (Liu, Whelen et al. 1999, Reeb, Lutzoni et al. 2004), respectively]. The ITS (MT939248), TEF-1α (MT946880), and RPB2 (MT946881) sequences of the BJ-22.3 isolate were deposited in GenBank. BLASTN analysis of these sequences showed over 99% nucleotide sequence identity with members of the FIESC: the ITS sequence showed 99.6% identity (544/546 bp) to F. lacertarum strain NRRL 20423 (GQ505682); the TEF-1α sequence showed 99.4% similarity (673/677 bp) to F. ipomoeae strain NRRL 43637 (GQ505664); and the RPB2 sequence showed 99.6% identity (1883/1901 bp) to F. equiseti strain GZUA.1657 (MG839492). Phylogenetic analysis using concatenated sequences of ITS, TEF-1α, and RPB2 showed that BJ-22.3 clustered monophyletically with strains of F. ipomoeae. Therefore, based on morphological and molecular characteristics, the isolate BJ-22.3 was identified as F. ipomoeae. To verify the F. ipomoeae isolate's pathogenicity, nine 1.5-year-old B. striata plants were inoculated with three 5 × 5 mm mycelial discs of strain BJ-22.3 from 4-day-old PDA cultures. Additionally, three control plants were inoculated with sterile PDA discs. The experiments were replicated three times. All plants were enclosed in transparent plastic bags and incubated in a greenhouse at 26°C for 14 days. Four days post-inoculation, leaf spot symptoms appeared on the inoculated leaves, while no symptoms were observed in control plants. Finally, F. ipomoeae was consistently re-isolated from leaf lesions from the infected plants. To our knowledge, this is the first report of F. ipomoeae causing leaf spot disease on B. striata in China. The spread of this disease might pose a serious threat to the production of B. striata. Growers should implement disease management to minimize the risks posed by this pathogen.