Multi-scale biomedical knowledge networks are expanding with emerging experimental technologies that generates multi-scale biomedical big data. Link prediction is increasingly used especially in bipartite biomedical networks to identify hidden biological interactions and relationshipts between key entities such as compounds, targets, gene and diseases. We propose a Graph Neural Networks (GNN) method, namely Graph Pair based Link Prediction model (GPLP), for predicting biomedical network links simply based on their topological interaction information. In GPLP, 1-hop subgraphs extracted from known network interaction matrix is learnt to predict missing links. To evaluate our method, three heterogeneous biomedical networks were used, i.e. Drug-Target Interaction network (DTI), Compound-Protein Interaction network (CPI) from NIH Tox21, and Compound-Virus Inhibition network (CVI). Our proposed GPLP method significantly outperforms over the state-of-the-art baselines. In addition, different network incompleteness is analysed with our devised protocol, and we also design an effective approach to improve the model robustness towards incomplete networks. Our method demonstrates the potential applications in other biomedical networks.
The University of Texas at Austin（得克萨斯大学奥斯汀分校）；AbbVie Inc.（艾伯维公司）；Indiana University（印第安纳大学）
Knowledge Graphs have been one of the fundamental methods for integrating heterogeneous data sources. Integrating heterogeneous data sources is crucial, especially in the biomedical domain, where central data-driven tasks such as drug discovery rely on incorporating information from different biomedical databases. These databases contain various biological entities and relations such as proteins (PDB), genes (Gene Ontology), drugs (DrugBank), diseases (DDB), and protein-protein interactions (BioGRID). The process of semantically integrating heterogeneous biomedical databases is often riddled w…
Hyperbolic Graph Representation Learning for Differential Diagnosis on Biomedical Knowledge Graphs
Università degli Studi di Milano（米兰大学）；Università della Svizzera italiana (USI)（瑞士意大利语大学）；Politecnico di Milano（米兰理工大学）；Aalto University（阿尔托大学）；FGV（杰图利奥·瓦加斯基金会）
University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
Biomedical entity linking is the task of linking entity mentions in a biomedical document to referent entities in a knowledge base. Recently, many BERT-based models have been introduced for the task. While these models have achieved competitive results on many datasets, they are computationally expensive and contain about 110M parameters. Little is known about the factors contributing to their impressive performance and whether the over-parameterization is needed. In this work, we shed some light on the inner working mechanisms of these large BERT-based models. Through a set of probing experim…
University of Cambridge（剑桥大学）；Monash University（莫纳什大学）；Amazon Alexa（亚马逊Alexa）
Despite the widespread success of self-supervised learning via masked language models (MLM), accurately capturing fine-grained semantic relationships in the biomedical domain remains a challenge. This is of paramount importance for entity-level tasks such as entity linking where the ability to model entity relations (especially synonymy) is pivotal. To address this challenge, we propose SapBERT, a pretraining scheme that self-aligns the representation space of biomedical entities. We design a scalable metric learning framework that can leverage UMLS, a massive collection of biomedical ontologi…
Global Health Drug Discovery Institute（全球健康药物研发中心）
The study of molecule-target interaction is quite important for drug discovery in terms of target identification, hit identification, pathway study, drug-drug interaction, etc. Most existing methodologies utilize either biomedical network information or molecule structural features to predict potential interaction link. However, the biomedical network information based methods usually suffer from cold start problem, while structure based methods often give limited performance due to the structure/interaction assumption and data quality. To address these issues, we propose a pseudo-siamese Grap…
Biomedical knowledge graphs (KG) are heterogenous networks consisting of biological entities as nodes and relations between them as edges. These entities and relations are extracted from millions of research papers and unified in a single resource. The goal of biomedical multi-hop question-answering over knowledge graph (KGQA) is to help biologist and scientist to get valuable insights by asking questions in natural language. Relevant answers can be found by first understanding the question and then querying the KG for right set of nodes and relationships to arrive at an answer. To model the q…
College of Computer Science and Electronic Engineering, Hunan University（湖南大学计算机科学与电子工程学院）；Institute of Open Source Software and Platform, Peng Cheng Laboratory（鹏城实验室开源软件与平台研究所）；Computer Network Information Center, Chinese Academy of Sciences（中国科学院计算机网络信息中心）；The State Key Laboratory of Chemo/Biosensing and Chemometrics, Hunan University（湖南大学化学生物传感与计量学国家重点实验室）
Self-supervised representation learning (SSL) on biomedical networks provides new opportunities for drug discovery. However, how to effectively combine multiple SSL models is still challenging and has been rarely explored. Therefore, we propose multi-task joint strategies of self-supervised representation learning on biomedical networks for drug discovery, named MSSL2drug. We design six basic SSL tasks inspired by various modality features including structures, semantics, and attributes in heterogeneous biomedical networks. Importantly, fifteen combinations of multiple tasks are evaluated by a…
The Ohio State University（俄亥俄州立大学）；The Research Institute at Nationwide Children’s Hospital（全国儿童医院研究所）；Huazhong Agricultural University（华中农业大学）
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 evalua…
CHAT: Beyond Contrastive Graph Transformer for Link Prediction in Heterogeneous Networks
Chinese Academy of Medical Sciences & Peking Union Medical College（中国医学科学院北京协和医学院）；Baidu Research（百度研究院）；Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））
Hypothesis generation is becoming a crucial time-saving technique which allows biomedical researchers to quickly discover implicit connections between important concepts. Typically, these systems operate on domain-specific fractions of public medical data. MOLIERE, in contrast, utilizes information from over 24.5 million documents. At the heart of our approach lies a multi-modal and multi-relational network of biomedical objects extracted from several heterogeneous datasets from the National Center for Biotechnology Information (NCBI). These objects include but are not limited to scientific pa…
Northeastern University（东北大学）；Microsoft Research Asia（微软亚洲研究院）；Institute of Computing Technology, Chinese Academy of Sciences（中国科学院计算技术研究所）
With the development of temporal networks such as E-commerce networks and social networks, the issue of temporal link prediction has attracted increasing attention in recent years. The Temporal Link Prediction task of WSDM Cup 2022 expects a single model that can work well on two kinds of temporal graphs simultaneously, which have quite different characteristics and data properties, to predict whether a link of a given type will occur between two given nodes within a given time span. Our team, named as nothing here, regards this task as a link prediction task in heterogeneous temporal networks…
The Australian National University（澳大利亚国立大学）；Alibaba Group（阿里巴巴集团）
Network embedding aims to learn low-dimensional representations of nodes while capturing structure information of networks. It has achieved great success on many tasks of network analysis such as link prediction and node classification. Most of existing network embedding algorithms focus on how to learn static homogeneous networks effectively. However, networks in the real world are more complex, e.g., networks may consist of several types of nodes and edges (called heterogeneous information) and may vary over time in terms of dynamic nodes and edges (called evolutionary patterns). Limited wor…
In this paper, we study network representation learning for tripartite heterogeneous networks which learns node representation features for networks with three types of node entities. We argue that tripartite networks are common in real world applications, and the essential challenge of the representation learning is the heterogeneous relations between various node types and links in the network. To tackle the challenge, we develop a tripartite heterogeneous network embedding called TriNE. The method considers unique user-item-tag tripartite relationships, to build an objective function to mod…
People usually get involved in multiple social networks to enjoy new services or to fulfill their needs. Many new social networks try to attract users of other existing networks to increase the number of their users. Once a user (called source user) of a social network (called source network) joins a new social network (called target network), a new inter-network link (called anchor link) is formed between the source and target networks. In this paper, we concentrated on predicting the formation of such anchor links between heterogeneous social networks. Unlike conventional link prediction pro…
The problem of ideology detection is to study the latent (political) placement for people, which is traditionally studied on politicians according to their voting behaviors. Recently, more and more studies begin to address the ideology detection problem for ordinary users based on their online behaviors that can be captured by social media, e.g., Twitter. As far as we are concerned, however, the vast majority of the existing methods on ideology detection on social media have oversimplified the problem as a binary classification problem (i.e., liberal vs. conservative). Moreover, though social …
Virginia Tech（弗吉尼亚理工大学）；Pacific Northwest National Laboratory（太平洋西北国家实验室）
Representation learning methods for heterogeneous networks produce a low-dimensional vector embedding for each node that is typically fixed for all tasks involving the node. Many of the existing methods focus on obtaining a static vector representation for a node in a way that is agnostic to the downstream application where it is being used. In practice, however, downstream tasks such as link prediction require specific contextual information that can be extracted from the subgraphs related to the nodes provided as input to the task. To tackle this challenge, we develop SLiCE, a framework brid…
Socio-technical systems usually consists of many intertwined networks, each connecting different types of objects (or actors) through a variety of means. As these networks are co-dependent, one can take advantage of this entangled structure to study interaction patterns in a particular network from the information provided by other related networks. A method is hence proposed and tested to recover the weights of missing or unobserved links in heterogeneous information networks (HIN) - abstract representations of systems composed of multiple types of entities and their relations. Given a pair o…
