Abstract: BACKGROUND: Link prediction is an important and well-studied problem in network biology. Recently, graph representation learning methods, including Graph Convolutional Network (GCN)-based node embedding have drawn increasing attention in link prediction. MOTIVATION: An important component of GCN-based network embedding is the convolution matrix, which is used to propagate features across the network. Existing algorithms use the degree-normalized adjacency matrix for this purpose, as this matrix is closely related to the graph Laplacian, capturing the spectral properties of the network. In parallel, it has been shown that GCNs with a single layer can generate more robust embeddings by reducing the number of parameters. Laplacian-based convolution is not well suited to single layered GCNs, as it limits the propagation of information to immediate neighbors of a node. RESULTS: Capitalizing on the rich literature on unsupervised link prediction, we propose using node similarity based convolution matrices in GCNs to compute node embeddings for link prediction. We consider eight representative node similarity measures (Common Neighbors, Jaccard Index, Adamic-Adar, Resource Allocation, Hub Depressed Index, Hub Promoted Index, Sorenson Index, Salton Index) for this purpose. We systematically compare the performance of the resulting algorithms against GCNs that use the degree-normalized adjacency matrix for convolution, as well as other link prediction algorithms. In our experiments, we use three link prediction tasks involving biomedical networks: drug-disease association (DDA) prediction, drug-drug interaction (DDI) prediction, protein-protein interaction (PPI) prediction. Our results show that node similarity-based convolution matrices significantly improve the link prediction performance of GCN-based embeddings. CONCLUSION: As sophisticated machine learning frameworks are increasingly employed in biological applications, historically well-established methods can be useful in making a head-start. AVAILABILITY: Our method, SiGraC, is implemented as a Python library and is freely available at https://github.com/mustafaCoskunAgu/SiGraC. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Abstract: BACKGROUND: Medulloblastoma (MB) is the most common malignant paediatric brain tumour and a leading cause of cancer-related mortality and morbidity. Existing treatment protocols are aggressive in nature resulting in significant neurological, intellectual and physical disabilities for the children undergoing treatment. Thus, there is an urgent need for improved, targeted therapies that minimize these harmful side effects. METHODS: We identified candidate drugs for MB using a network-based systems-pharmacogenomics approach: based on results from a functional genomics screen, we identified a network of interactions implicated in human MB growth regulation. We then integrated drugs and their known mechanisms of action, along with gene expression data from a large collection of medulloblastoma patients to identify drugs with potential to treat MB. RESULTS: Our analyses identified drugs targeting CDK4, CDK6 and AURKA as strong candidates for MB; all of these genes are well validated as drug targets in other tumour types. We also identified non-WNT MB as a novel indication for drugs targeting TUBB, CAD, SNRPA, SLC1A5, PTPRS, P4HB and CHEK2. Based upon these analyses, we subsequently demonstrated that one of these drugs, the new microtubule stabilizing agent, ixabepilone, blocked tumour growth in vivo in mice bearing patient-derived xenograft tumours of the Sonic Hedgehog and Group 3 subtype, providing the first demonstration of its efficacy in MB. CONCLUSIONS: Our findings confirm that this data-driven systems pharmacogenomics strategy is a powerful approach for the discovery and validation of novel therapeutic candidates relevant to MB treatment, and along with data validating ixabepilone in PDX models of the two most aggressive subtypes of medulloblastoma, we present the network analysis framework as a resource for the field.

Abstract: The incidence of various types of cancer is increasing globally. To reduce the critical side effects of cancer chemotherapy, naturally derived compounds have been considered for cancer treatment. Gymnosperms are a group of plants found worldwide that have traditionally been used for therapeutic applications. Paclitaxel is a commercially available anticancer drug derived from gymnosperms. Other natural compounds with anticancer activities, such as pinostrobin and pinocembrin, are extracted from pine heartwood, and pycnogenol and enzogenol from pine bark. Gymnosperms have great potential for further study for the discovery of new anticancer compounds. This review aims to provide a rational understanding and the latest developments in potential anticancer compounds derived from gymnosperms.

Abstract: Backgroud: The prediction of drug-target interactions (DTIs) is of great significance in drug development. It is time-consuming and expensive in traditional experimental methods. Machine learning can reduce the cost of prediction and is limited by the characteristics of imbalanced datasets and problems of essential feature selection. METHODS: The prediction method based on the Ensemble model of Multiple Feature Pairs (Ensemble-MFP) is introduced. Firstly, three negative sets are generated according to the Euclidean distance of three feature pairs. Then, the negative samples of the validation set/test set are randomly selected from the union set of the three negative sets in the validation set/test set. At the same time, the ensemble model with weight is optimized and applied to the test set. RESULTS: The area under the receiver operating characteristic curve (area under ROC, AUC) in three out of four sub-datasets in gold standard datasets was more than 94.0% in the prediction of new drugs. The effectiveness of the proposed method is also shown with the comparison of state-of-the-art methods and demonstration of predicted drug-target pairs. CONCLUSION: The Ensemble-MFP can weigh the existing feature pairs and has a good prediction effect for general prediction on new drugs.

Abstract: Identification of small molecules is a critical task in various areas of life science. Recent advances in mass spectrometry have enabled the collection of tandem mass spectra of small molecules from hundreds of thousands of environments. To identify which molecules are present in a sample, one can search mass spectra collected from the sample against millions of molecular structures in small molecule databases. The existing approaches are based on chemistry domain knowledge, and they fail to explain many of the peaks in mass spectra of small molecules. Here, we present molDiscovery, a mass spectral database search method that improves both efficiency and accuracy of small molecule identification by learning a probabilistic model to match small molecules with their mass spectra. A search of over 8 million spectra from the Global Natural Product Social molecular networking infrastructure shows that molDiscovery correctly identify six times more unique small molecules than previous methods.

Abstract: The antibiotic combination trimethoprim (TMP)-sulfamethoxazole (SMX) has a broad spectrum of activity and is used for the treatment of numerous infections, but pediatric pharmacokinetic (PK) data are limited. We previously published population PK (popPK) models of oral TMP-SMX in pediatric patients based on sparse opportunistically collected data (POPS study) (J. Autmizguine, C. Melloni, C. P. Hornik, S. Dallefeld, et al., Antimicrob Agents Chemother 62:e01813-17, 2017, https://doi.org/10.1128/AAC.01813-17). We performed a separate PK study of oral TMP-SMX in infants and children with more-traditional PK sample collection and independently developed new popPK models of TMP-SMX using this external data set. The POPS data set and the external data set were each used to evaluate both popPK models. The external TMP model had a model and error structure identical to those of the POPS TMP model, with typical values for PK parameters within 20%. The external SMX model did not identify the covariates in the POPS SMX model as significant. The external popPK models predicted higher exposures to TMP (median overprediction of 0.13 mg/liter for the POPS data set and 0.061 mg/liter for the external data set) and SMX (median overprediction of 1.7 mg/liter and 0.90 mg/liter) than the POPS TMP (median underprediction of 0.016 mg/liter and 0.39 mg/liter) and SMX (median underprediction of 1.2 mg/liter and 14 mg/liter) models. Nonetheless, both models supported TMP-SMX dose increases in infants and young children for resistant pathogens with a MIC of 1 mg/liter, although the required dose increase based on the external model was lower. (The POPS and external studies have been registered at ClinicalTrials.gov under registration no. NCT01431326 and NCT02475876, respectively.).

Abstract: Existing evidence is inconclusive whether meropenem dosing should be adjusted in patients receiving extracorporeal membrane oxygenation (ECMO). Therefore, the aim of this observational matched cohort study was to evaluate the effect of ECMO on pharmacokinetic (PK) variability and target attainment (TA) of meropenem. Patients admitted to the intensive care unit (ICU) simultaneously treated with meropenem and ECMO were eligible. Patients were matched 1:1, based on renal function and body weight, with non-ECMO ICU patients. Meropenem blood sampling was performed over one or two dosing intervals. Population PK modelling was performed using NONMEM7.5. TA was defined as free meropenem concentrations >2 or 8 mg/L (i.e., 1 or 4x minimal inhibitory concentration, respectively) throughout the whole dosing interval. In total, 25 patients were included, contributing 27 dosing intervals. The overall TA was 56% and 26% for the 2 mg/L and 8 mg/L target, respectively. Population PK modelling identified estimated glomerular filtration rate according to the Chronic Kidney Disease Epidemiology equation and body weight, but not ECMO, as significant predictors. In conclusion, TA of meropenem was confirmed to be poor under standard dosing in critically ill patients but was not found to be influenced by ECMO. Future studies should focus on applying dose optimisation strategies for meropenem based on renal function, regardless of ECMO.

Abstract: Previous studies indicated that miRNA plays an important role in human biological processes especially in the field of diseases. However, constrained by biotechnology, only a small part of the miRNA-disease associations has been verified by biological experiment. This impel that more and more researchers pay attention to develop efficient and high-precision computational methods for predicting the potential miRNA-disease associations. Based on the assumption that molecules are related to each other in human physiological processes, we developed a novel structural deep network embedding model (SDNE-MDA) for predicting miRNA-disease association using molecular associations network. Specifically, the SDNE-MDA model first integrating miRNA attribute information by Chao Game Representation (CGR) algorithm and disease attribute information by disease semantic similarity. Secondly, we extract feature by structural deep network embedding from the heterogeneous molecular associations network. Then, a comprehensive feature descriptor is constructed by combining attribute information and behavior information. Finally, Convolutional Neural Network (CNN) is adopted to train and classify these feature descriptors. In the five-fold cross validation experiment, SDNE-MDA achieved AUC of 0.9447 with the prediction accuracy of 87.38% on the HMDD v3.0 dataset. To further verify the performance of SDNE-MDA, we contrasted it with different feature extraction models and classifier models. Moreover, the case studies with three important human diseases, including Breast Neoplasms, Kidney Neoplasms, Lymphoma were implemented by the proposed model. As a result, 47, 46 and 46 out of top-50 predicted disease-related miRNAs have been confirmed by independent databases. These results anticipate that SDNE-MDA would be a reliable computational tool for predicting potential miRNA-disease associations.

Abstract: The number of candidate molecules for new non-narcotic analgesics is extremely limited. Here, we report the identification of thiowurtzine, a new potent analgesic molecule with promising application in chronic pain treatment. We describe the chemical synthesis of this unique compound derived from the hexaazaisowurtzitane (CL-20) explosive molecule. Then, we use animal experiments to assess its analgesic activity in vivo upon chemical, thermal, and mechanical exposures, compared to the effect of several reference drugs. Finally, we investigate the potential receptors of thiowurtzine in order to better understand its complex mechanism of action. We use docking, molecular modeling, and molecular dynamics simulations to identify and characterize the potential targets of the drug and confirm the results of the animal experiments. Our findings finally indicate that thiowurtzine may have a complex mechanism of action by essentially targeting the mu opioid receptor, the TRPA1 ion channel, and the Cav voltage-gated calcium channel.

Abstract: Many of genes mediating Known Drug-Disease Association (KDDA) are escaped from experimental detection. Identifying of these genes (hidden genes) is of great significance for understanding disease pathogenesis and guiding drug repurposing. Here, we presented a novel computational tool, called KDDANet, for systematic and accurate uncovering the hidden genes mediating KDDA from the perspective of genome-wide functional gene interaction network. KDDANet demonstrated the competitive performances in both sensitivity and specificity of identifying genes in mediating KDDA in comparison to the existing state-of-the-art methods. Case studies on Alzheimer's disease (AD) and obesity uncovered the mechanistic relevance of KDDANet predictions. Furthermore, when applied with multiple types of cancer-omics datasets, KDDANet not only recapitulated known genes mediating KDDAs related to cancer, but also revealed novel candidates that offer new biological insights. Importantly, KDDANet can be used to discover the shared genes mediating multiple KDDAs. KDDANet can be accessed at http://www.kddanet.cn and the code can be freely downloaded at https://github.com/huayu1111/KDDANet .