Abstract: MOTIVATION: Graph embedding learning that aims to automatically learn low-dimensional node representations, has drawn increasing attention in recent years. To date, most recent graph embedding methods are evaluated on social and information networks and are not comprehensively studied on biomedical networks under systematic experiments and analyses. On the other hand, for a variety of biomedical network analysis tasks, traditional techniques such as matrix factorization (which can be seen as a type of graph embedding methods) have shown promising results, and hence there is a need to systematically evaluate the more recent graph embedding methods (e.g. random walk-based and neural network-based) in terms of their usability and potential to further the state-of-the-art. RESULTS: We select 11 representative graph embedding methods and conduct a systematic comparison on 3 important biomedical link prediction tasks: drug-disease association (DDA) prediction, drug-drug interaction (DDI) prediction, protein-protein interaction (PPI) prediction; and 2 node classification tasks: medical term semantic type classification, protein function prediction. Our experimental results demonstrate that the recent graph embedding methods achieve promising results and deserve more attention in the future biomedical graph analysis. Compared with three state-of-the-art methods for DDAs, DDIs and protein function predictions, the recent graph embedding methods achieve competitive performance without using any biological features and the learned embeddings can be treated as complementary representations for the biological features. By summarizing the experimental results, we provide general guidelines for properly selecting graph embedding methods and setting their hyper-parameters for different biomedical tasks. AVAILABILITY AND IMPLEMENTATION: As part of our contributions in the paper, we develop an easy-to-use Python package with detailed instructions, BioNEV, available at: https://github.com/xiangyue9607/BioNEV, including all source code and datasets, to facilitate studying various graph embedding methods on biomedical tasks. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Abstract: Integration of multiple sources and data levels provides a great insight into the complex associations between human and malaria systems. In this study, a meta-analysis framework was developed based on a heterogeneous network model for integrating human-malaria protein similarities, a human protein interaction network, and a Plasmodium vivax protein interaction network. An iterative network propagation was performed on the heterogeneous network until we obtained stabilized weights. The association scores were calculated for qualifying a novel potential human-malaria protein association. This method provided a better performance compared to random experiments. After that, the stabilized network was clustered into association modules. The potential association candidates were then thoroughly analyzed by statistical enrichment analysis with protein complexes and known drug targets. The most promising target proteins were the succinate dehydrogenase protein complex in the human citrate (TCA) cycle pathway and the nicotinic acetylcholine receptor in the human central nervous system. Promising associations and potential drug targets were also provided for further studies and designs in therapeutic approaches for malaria at a systematic level. In conclusion, this method is efficient to identify new human-malaria protein associations and can be generalized to infer other types of association studies to further advance biomedical science.

Abstract: Single-pass membrane receptors contain extracellular domains that respond to external stimuli and transmit information to intracellular domains through a single transmembrane (TM) alpha-helix. Because membrane receptors have various roles in homeostasis, signaling malfunctions of these receptors can cause disease. Despite their importance, there is still much to be understood mechanistically about how single-pass receptors are activated. In general, single-pass receptors respond to extracellular stimuli via alterations in their oligomeric state. The details of this process are still the focus of intense study, and several lines of evidence indicate that the TM domain (TMD) of the receptor plays a central role. We discuss three major mechanistic hypotheses for receptor activation: ligand-induced dimerization, ligand-induced rotation, and receptor clustering. Recent observations suggest that receptors can use a combination of these activation mechanisms and that technical limitations can bias interpretation. Short peptides derived from receptor TMDs, which can be identified by screening or rationally developed on the basis of the structure or sequence of their targets, have provided critical insights into receptor function. Here, we explore recent evidence that, depending on the target receptor, TMD peptides cannot only inhibit but also activate target receptors and can accommodate novel, bifunctional designs. Furthermore, we call for more sharing of negative results to inform the TMD peptide field, which is rapidly transforming into a suite of unique tools with the potential for future therapeutics.

Abstract: Few therapeutic methods exist for preventing preterm birth (PTB), or delivery before completing 37 weeks of gestation. In the US, progesterone (P4) supplementation is the only FDA-approved drug for use in preventing recurrent spontaneous PTB. However, P4 has limited effectiveness, working in only approximately one-third of cases. Computational drug repositioning leverages data on existing drugs to discover novel therapeutic uses. We used a rank-based pattern-matching strategy to compare the differential gene expression signature for PTB to differential gene expression drug profiles in the Connectivity Map database and assigned a reversal score to each PTB-drug pair. Eighty-three drugs, including P4, had significantly reversed differential gene expression compared with that found for PTB. Many of these compounds have been evaluated in the context of pregnancy, with 13 belonging to pregnancy category A or B - indicating no known risk in human pregnancy. We focused our validation efforts on lansoprazole, a proton-pump inhibitor, which has a strong reversal score and a good safety profile. We tested lansoprazole in an animal inflammation model using LPS, which showed a significant increase in fetal viability compared with LPS treatment alone. These promising results demonstrate the effectiveness of the computational drug repositioning pipeline to identify compounds that could be effective in preventing PTB.

Abstract: The chemical sciences are producing an unprecedented amount of large, high-dimensional data sets containing chemical structures and associated properties. However, there are currently no algorithms to visualize such data while preserving both global and local features with a sufficient level of detail to allow for human inspection and interpretation. Here, we propose a solution to this problem with a new data visualization method, TMAP, capable of representing data sets of up to millions of data points and arbitrary high dimensionality as a two-dimensional tree (http://tmap.gdb.tools). Visualizations based on TMAP are better suited than t-SNE or UMAP for the exploration and interpretation of large data sets due to their tree-like nature, increased local and global neighborhood and structure preservation, and the transparency of the methods the algorithm is based on. We apply TMAP to the most used chemistry data sets including databases of molecules such as ChEMBL, FDB17, the Natural Products Atlas, DSSTox, as well as to the MoleculeNet benchmark collection of data sets. We also show its broad applicability with further examples from biology, particle physics, and literature.

Abstract: Chemogenomics, also called proteochemometrics, covers a range of computational methods that can be used to predict protein-ligand interactions at large scales in the protein and chemical spaces. They differ from more classical ligand-based methods (also called QSAR) that predict ligands for a given protein receptor. In the context of drug discovery process, chemogenomics allows to tackle the question of predicting off-target proteins for drug candidates, one of the main causes of undesirable side-effects and failure within drugs development processes. The present study compares shallow and deep machine-learning approaches for chemogenomics, and explores data augmentation techniques for deep learning algorithms in chemogenomics. Shallow machine-learning algorithms rely on expert-based chemical and protein descriptors, while recent developments in deep learning algorithms enable to learn abstract numerical representations of molecular graphs and protein sequences, in order to optimise the performance of the prediction task. We first propose a formulation of chemogenomics with deep learning, called the chemogenomic neural network (CN), as a feed-forward neural network taking as input the combination of molecule and protein representations learnt by molecular graph and protein sequence encoders. We show that, on large datasets, the deep learning CN model outperforms state-of-the-art shallow methods, and competes with deep methods with expert-based descriptors. However, on small datasets, shallow methods present better prediction performance than deep learning methods. Then, we evaluate data augmentation techniques, namely multi-view and transfer learning, to improve the prediction performance of the chemogenomic neural network. We conclude that a promising research direction is to integrate heterogeneous sources of data such as auxiliary tasks for which large datasets are available, or independently, multiple molecule and protein attribute views.

Abstract: Aim: High-throughput phenotypic screens have emerged as a promising avenue for small-molecule drug discovery. The challenge faced in high-throughput phenotypic screens is target deconvolution once a small molecule hit is identified. Chemogenomics libraries have emerged as an important tool for meeting this challenge. Here, we investigate their target-specificity by deriving a 'polypharmacology index' for broad chemogenomics screening libraries. Methods: All known targets of all the compounds in each library were plotted as a histogram and fitted to a Boltzmann distribution, whose linearized slope is indicative of the overall polypharmacology of the library. Results & conclusion: Comparison of libraries clearly distinguished the most target-specific library, which might be assumed to be more useful for target deconvolution in a phenotypic screen.

Abstract: The activation of the PI3K/AKT/mTOR pathway is a main driver of cell growth, proliferation, survival, and chemoresistance of cancer cells, and, for this reason, represents an attractive target for developing targeted anti-cancer drugs. There are plenty of preclinical data sustaining the anti-tumor activity of dual PI3K/mTOR inhibitors as single agents and in combination in lymphomas. Clinical responses, including complete remissions (especially in follicular lymphoma patients), are also observed in the very few clinical studies performed in patients that are affected by relapsed/refractory lymphomas or chronic lymphocytic leukemia. In this review, we summarize the literature on dual PI3K/mTOR inhibitors focusing on the lymphoma setting, presenting both the three compounds still in clinical development and those with a clinical program stopped or put on hold.

Abstract: Tuberculosis (TB) remains a serious threat to global public health, responsible for an estimated 1.5 million mortalities in 2018. While there are available therapeutics for this infection, slow-acting drugs, poor patient compliance, drug toxicity, and drug resistance require the discovery of novel TB drugs. Discovering new and more potent antibiotics that target novel TB protein targets is an attractive strategy towards controlling the global TB epidemic. In silico strategies can be applied at multiple stages of the drug discovery paradigm to expedite the identification of novel anti-TB therapeutics. In this paper, we discuss the current TB treatment, emergence of drug resistance, and the effective application of computational tools to the different stages of TB drug discovery when combined with traditional biochemical methods. We will also highlight the strengths and points of improvement in in silico TB drug discovery research, as well as possible future perspectives in this field.

Abstract: PURPOSE: In this work, we introduce CDGnet (Cancer-Drug-Gene Network), an evidence-based network approach for recommending targeted cancer therapies. CDGnet represents a user-friendly informatics tool that expands the range of targeted therapy options for patients with cancer who undergo molecular profiling by including the biologic context via pathway information. METHODS: CDGnet considers biologic pathway information specifically by looking at targets or biomarkers downstream of oncogenes and is personalized for individual patients via user-inputted molecular alterations and cancer type. It integrates a number of different sources of knowledge: patient-specific inputs (molecular alterations and cancer type), US Food and Drug Administration-approved therapies and biomarkers (curated from DailyMed), pathways for specific cancer types (from Kyoto Encyclopedia of Genes and Genomes [KEGG]), gene-drug connections (from DrugBank), and oncogene information (from KEGG). We consider 4 different evidence-based categories for therapy recommendations. Our tool is delivered via an R/Shiny Web application. For the 2 categories that use pathway information, we include an interactive Sankey visualization built on top of d3.js that also provides links to PubChem. RESULTS: We present a scenario for a patient who has estrogen receptor (ER)-positive breast cancer with FGFR1 amplification. Although many therapies exist for patients with ER-positive breast cancer, FGFR1 amplifications may confer resistance to such treatments. CDGnet provides therapy recommendations, including PIK3CA, MAPK, and RAF inhibitors, by considering targets or biomarkers downstream of FGFR1. CONCLUSION: CDGnet provides results in a number of easily accessible and usable forms, separating targeted cancer therapies into categories in an evidence-based manner that incorporates biologic pathway information.