GEval: A Modular and Extensible Evaluation Framework for Graph Embedding Techniques
Ontology highlight
ABSTRACT:
SUBMITTER: Harth A
PROVIDER: S-EPMC7250612 | biostudies-literature | 2020 May
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:In silico prediction of drug-target interactions is a critical phase in the sustainable drug development process, especially when the research focus is to capitalize on the repositioning of existing drugs. However, developing such computational methods is not an easy task, but is much needed, as current methods that predict potential drug-target interactions suffer from high false-positive rates. Here we introduce DTiGEMS+, a computational method that predicts Drug-Target interactions using Graph Embedding, graph Mining, and Similarity-based techniques. DTiGEMS+?combines similarity-based as well as feature-based approaches, and models the identification of novel drug-target interactions as a link prediction problem in a heterogeneous network. DTiGEMS+?constructs the heterogeneous network by augmenting the known drug-target interactions graph with two other complementary graphs namely: drug-drug similarity, target-target similarity. DTiGEMS+?combines different computational techniques to provide the final drug target prediction, these techniques include graph embeddings, graph mining, and machine learning. DTiGEMS+?integrates multiple drug-drug similarities and target-target similarities into the final heterogeneous graph construction after applying a similarity selection procedure as well as a similarity fusion algorithm. Using four benchmark datasets, we show DTiGEMS+?substantially improves prediction performance compared to other state-of-the-art in silico methods developed to predict of drug-target interactions by achieving the highest average AUPR across all datasets (0.92), which reduces the error rate by 33.3% relative to the second-best performing model in the state-of-the-art methods comparison.
| S-EPMC7325230 | biostudies-literature
Project description:Here we study the semantic search and retrieval problem in biomedical digital libraries. First, we introduce MedGraph, a knowledge graph embedding-based method that provides semantic relevance retrieval and ranking for the biomedical literature indexed in PubMed. Second, we evaluate our approach using PubMed's Best Match algorithm. Moreover, we compare our method MedGraph to a traditional TF-IDF-based algorithm. Third, we use a dataset extracted from PubMed, including 30 million articles' metadata such as abstracts, author information, citation information, and extracted biological entity mentions. We pull a subset of the dataset to evaluate MedGraph using predefined queries with ground truth ranked results. To our knowledge, this technique has not been explored before in biomedical information retrieval. In addition, our results provide some evidence that semantic approaches to search and relevance in biomedical digital libraries that rely on knowledge graph modeling offer better search relevance results when compared with traditional methods in terms of objective metrics.
| S-EPMC9627348 | biostudies-literature
Project description:In this research, we improve upon the current state of the art in entity retrieval by re-ranking the result list using graph embeddings. The paper shows that graph embeddings are useful for entity-oriented search tasks. We demonstrate empirically that encoding information from the knowledge graph into (graph) embeddings contributes to a higher increase in effectiveness of entity retrieval results than using plain word embeddings. We analyze the impact of the accuracy of the entity linker on the overall retrieval effectiveness. Our analysis further deploys the cluster hypothesis to explain the observed advantages of graph embeddings over the more widely used word embeddings, for user tasks involving ranking entities.
| S-EPMC7148245 | biostudies-literature
Project description:In bioinformatics, it is important to build extensible and low-maintenance systems that are able to deal with the new tools and data formats that are constantly being developed. The traditional and simplest implementation of pipelines involves hardcoding the execution steps into programs or scripts. This approach can lead to problems when a pipeline is expanding because the incorporation of new tools is often error prone and time consuming. Current approaches to pipeline development such as workflow management systems focus on analysis tasks that are systematically repeated without significant changes in their course of execution, such as genome annotation. However, more dynamism on the pipeline composition is necessary when each execution requires a different combination of steps.We propose a graph-based approach to implement extensible and low-maintenance pipelines that is suitable for pipeline applications with multiple functionalities that require different combinations of steps in each execution. Here pipelines are composed automatically by compiling a specialised set of tools on demand, depending on the functionality required, instead of specifying every sequence of tools in advance. We represent the connectivity of pipeline components with a directed graph in which components are the graph edges, their inputs and outputs are the graph nodes, and the paths through the graph are pipelines. To that end, we developed special data structures and a pipeline system algorithm. We demonstrate the applicability of our approach by implementing a format conversion pipeline for the fields of population genetics and genetic epidemiology, but our approach is also helpful in other fields where the use of multiple software is necessary to perform comprehensive analyses, such as gene expression and proteomics analyses. The project code, documentation and the Java executables are available under an open source license at http://code.google.com/p/dynamic-pipeline. The system has been tested on Linux and Windows platforms.Our graph-based approach enables the automatic creation of pipelines by compiling a specialised set of tools on demand, depending on the functionality required. It also allows the implementation of extensible and low-maintenance pipelines and contributes towards consolidating openness and collaboration in bioinformatics systems. It is targeted at pipeline developers and is suited for implementing applications with sequential execution steps and combined functionalities. In the format conversion application, the automatic combination of conversion tools increased both the number of possible conversions available to the user and the extensibility of the system to allow for future updates with new file formats.
| S-EPMC3496580 | biostudies-other
Project description:Biologists and biochemists have at their disposal a number of excellent, publicly available data resources such as UniProt, KEGG, and NCBI Taxonomy, which catalogue biological entities. Despite the usefulness of these resources, they remain fundamentally unconnected. While links may appear between entries across these databases, users are typically only able to follow such links by manual browsing or through specialised workflows. Although many of the resources provide web-service interfaces for computational access, performing federated queries across databases remains a non-trivial but essential activity in interdisciplinary systems and synthetic biology programmes. What is needed are integrated repositories to catalogue both biological entities and-crucially-the relationships between them. Such a resource should be extensible, such that newly discovered relationships-for example, those between novel, synthetic enzymes and non-natural products-can be added over time. With the introduction of graph databases, the barrier to the rapid generation, extension and querying of such a resource has been lowered considerably. With a particular focus on metabolic engineering as an illustrative application domain, biochem4j, freely available at http://biochem4j.org, is introduced to provide an integrated, queryable database that warehouses chemical, reaction, enzyme and taxonomic data from a range of reliable resources. The biochem4j framework establishes a starting point for the flexible integration and exploitation of an ever-wider range of biological data sources, from public databases to laboratory-specific experimental datasets, for the benefit of systems biologists, biosystems engineers and the wider community of molecular biologists and biological chemists.
| S-EPMC5510799 | biostudies-literature
Project description:Many scientific datasets are of high dimension, and the analysis usually requires retaining the most important structures of data. Principal curve is a widely used approach for this purpose. However, many existing methods work only for data with structures that are mathematically formulated by curves, which is quite restrictive for real applications. A few methods can overcome the above problem, but they either require complicated human-made rules for a specific task with lack of adaption flexibility to different tasks, or cannot obtain explicit structures of data. To address these issues, we develop a novel principal graph and structure learning framework that captures the local information of the underlying graph structure based on reversed graph embedding. As showcases, models that can learn a spanning tree or a weighted undirected `1 graph are proposed, and a new learning algorithm is developed that learns a set of principal points and a graph structure from data, simultaneously. The new algorithm is simple with guaranteed convergence. We then extend the proposed framework to deal with large-scale data. Experimental results on various synthetic and six real world datasets show that the proposed method compares favorably with baselines and can uncover the underlying structure correctly.
| S-EPMC5899072 | biostudies-literature
Project description:BackgroundCurrent approaches to identifying drug-drug interactions (DDIs), include safety studies during drug development and post-marketing surveillance after approval, offer important opportunities to identify potential safety issues, but are unable to provide complete set of all possible DDIs. Thus, the drug discovery researchers and healthcare professionals might not be fully aware of potentially dangerous DDIs. Predicting potential drug-drug interaction helps reduce unanticipated drug interactions and drug development costs and optimizes the drug design process. Methods for prediction of DDIs have the tendency to report high accuracy but still have little impact on translational research due to systematic biases induced by networked/paired data. In this work, we aimed to present realistic evaluation settings to predict DDIs using knowledge graph embeddings. We propose a simple disjoint cross-validation scheme to evaluate drug-drug interaction predictions for the scenarios where the drugs have no known DDIs.ResultsWe designed different evaluation settings to accurately assess the performance for predicting DDIs. The settings for disjoint cross-validation produced lower performance scores, as expected, but still were good at predicting the drug interactions. We have applied Logistic Regression, Naive Bayes and Random Forest on DrugBank knowledge graph with the 10-fold traditional cross validation using RDF2Vec, TransE and TransD. RDF2Vec with Skip-Gram generally surpasses other embedding methods. We also tested RDF2Vec on various drug knowledge graphs such as DrugBank, PharmGKB and KEGG to predict unknown drug-drug interactions. The performance was not enhanced significantly when an integrated knowledge graph including these three datasets was used.ConclusionWe showed that the knowledge embeddings are powerful predictors and comparable to current state-of-the-art methods for inferring new DDIs. We addressed the evaluation biases by introducing drug-wise and pairwise disjoint test classes. Although the performance scores for drug-wise and pairwise disjoint seem to be low, the results can be considered to be realistic in predicting the interactions for drugs with limited interaction information.
| S-EPMC6921491 | biostudies-literature
Project description:Recent advances in chromatin architecture profiling technologies, such as single-cell Hi-C (scHi-C), allow us to dissect the heterogeneity of chromosome higher-order structures across diverse cell states and different individuals. However, scHi-C experiments are still expensive and not immediately available for population-scale profiling. Here, we present scENCORE, a computational method, to reconstruct personalized and cell-type-specific higher-order chromatin structures, such as A/B compartments, directly from more cost-effective and widely available single-cell epigenetic data (e.g., scATAC-seq). We apply scENCORE on scATAC-seq data from post-mortem prefrontal cortex brains and demonstrate its utility to 1) project Mega-base scale chromatin regions into lower dimensional space by leveraging graph embedding technologies based on cell-type-specific co-variability patterns, 2) define A/B compartments via unsupervised clustering, 3) perform an alignment algorithm for multi-graph embedding to derive comparable chromatin representations and highlight dynamic chromatin compartments across cell states and individuals. Validated by Hi-C experiments using FACS-sorted cells, scENCORE can faithfully reconstruct cell-type-specific chromatin compartments. Furthermore, scENCORE uniformly constructs chromosome conformation across population-scale scATAC-seq data and discovers key 3D structural switching events associated with psychiatric disorders. In summary, scENCORE allows cost-effective cell-type-specific and personalized reconstruction that delineate higher-order chromatin structures.
2023-09-27 | GSE216270 | GEO
Project description:Graph embedding methods are useful for a wide range of graph analysis tasks including link prediction and node classification. Most graph embedding methods learn only the topological structure of graphs. Nevertheless, it has been shown that the incorporation of node attributes is beneficial in improving the expressive power of node embeddings. However, real-world graphs are often noisy in terms of structure and/or attributes (missing and/or erroneous edges/attributes). Most existing graph embedding methods are susceptible to this noise, as they do not consider uncertainty during the modelling process. In this paper, we introduce RASE, a Robust Attribute and Structure preserving graph Embedding model. RASE is a novel graph representation learning model which effectively preserves both graph structure and node attributes through a unified loss function. To be robust, RASE uses a denoising attribute auto-encoder to deal with node attribute noise, and models uncertainty in the embedding space as Gaussians to cope with graph structure noise. We evaluate the performance of RASE through an extensive experimental study on various real-world datasets. Results demonstrate that RASE outperforms state-of-the-art embedding methods on multiple graph analysis tasks and is robust to both structure and attribute noise.
| S-EPMC7206273 | biostudies-literature
Project description:BackgroundProtein-protein interactions (PPIs) are central to many biological processes. Considering that the experimental methods for identifying PPIs are time-consuming and expensive, it is important to develop automated computational methods to better predict PPIs. Various machine learning methods have been proposed, including a deep learning technique which is sequence-based that has achieved promising results. However, it only focuses on sequence information while ignoring the structural information of PPI networks. Structural information of PPI networks such as their degree, position, and neighboring nodes in a graph has been proved to be informative in PPI prediction.ResultsFacing the challenge of representing graph information, we introduce an improved graph representation learning method. Our model can study PPI prediction based on both sequence information and graph structure. Moreover, our study takes advantage of a representation learning model and employs a graph-based deep learning method for PPI prediction, which shows superiority over existing sequence-based methods. Statistically, Our method achieves state-of-the-art accuracy of 99.15% on Human protein reference database (HPRD) dataset and also obtains best results on Database of Interacting Protein (DIP) Human, Drosophila, Escherichia coli (E. coli), and Caenorhabditis elegans (C. elegan) datasets.ConclusionHere, we introduce signed variational graph auto-encoder (S-VGAE), an improved graph representation learning method, to automatically learn to encode graph structure into low-dimensional embeddings. Experimental results demonstrate that our method outperforms other existing sequence-based methods on several datasets. We also prove the robustness of our model for very sparse networks and the generalization for a new dataset that consists of four datasets: HPRD, E.coli, C.elegan, and Drosophila.
| S-EPMC7372763 | biostudies-literature