COVID-19 Literature Knowledge Graph Construction and Drug Repurposing Report Generation

Abstract

To combat COVID-19, both clinicians and scientists need to digest vast amounts of relevant biomedical knowledge in scientific literature to understand the disease mechanism and related biological functions. We have developed a novel and comprehensive knowledge discovery framework, COVID-KG to extract finegrained multimedia knowledge elements (entities and their visual chemical structures, relations and events) from scientific literature. We then exploit the constructed multimedia knowledge graphs (KGs) for question answering and report generation, using drug repurposing as a case study. Our framework also provides detailed contextual sentences, subfigures, and knowledge subgraphs as evidence. All of the data, KGs, reports 1 , resources, and shared services are publicly available 2 .

Data science approaches to confronting the COVID-19 pandemic: a narrative review

“…Pham et al proposed a deep learning method, namely DeepCE, to model substructure-gene and gene-gene associations for predicting the differential gene expression profile perturbed by de novo chemicals, and demonstrated that DeepCE outperformed state-of-the-art, and could be applied to COVID-19 drug repurposing of COVID-19 with clinical evidence [63]. Zhou et al provided a useful review and helpful illustrations of these machine learning, and AI techniques for COVID-19 drug repurposing [91] The knowledge graph does not have to be manually constructed, except for the existing biological datasets, as machine learning and natural language processing (NLP) techniques are appropriate tools to automatically construct knowledge graphs from scientific literature [65].…”

“…More specifically, Chen et al combined the CORD-19 dataset [64] and the PubMed dataset [73] to identify COVID-19-related experts and bio-entities [69]. Another example is the COVID-KG framework, which could extract fine-grained multimedia knowledge elements from scientific literature [65]. The resulted knowledge is available at http://blender.cs.…”

Elsevier OA CC-By Corpus

“…As this dataset was designed for a particular cause, it is biased towards the biological and medial fields. The research output of the COVID19 dataset is extensive, including document summerisation [7], drug re-purposing [8] and risk factor identification [9]. A comprehensive overview will take a review article in itself to summarise this contribution.…”

Prostate cancer management with lifestyle intervention: From knowledge graph to Chatbot

“…There exist several representatives and comprehensive KGs, such as IBM Watson, 22 SNOMED‐CT, 23 and CmeKG, 24 and so forth, most of which aimed to save time and vigour, alleviate the pressure of physicians, and improve the accuracy of diagnosis to some extent. By investigation, the existing professional medical KGs mainly focus on diseases, 25 drugs, 26 cells, 27 literature, 28,29 proteins, genes, and organs 28 . However, personal lifestyles such as diet, sleep, vitamins, and environment, which are critical triggers of diseases, such as prostate cancer (PCa), can be an inspiration for novel direction in this field 30 …”