Article
Biochemical Research Methods
Guy Shtar, Asnat Greenstein-Messica, Eyal Mazuz, Lior Rokach, Bracha Shapira
Summary: This study proposes an adjacency biomedical text embedding (ABTE) method that combines known drug interactions and drug's biomedical text embeddings to predict both new and known drug interactions. The results demonstrate the superiority of this approach compared to other existing drug interaction prediction models and matrix factorization-based approaches. Furthermore, the study explores the use of different text embedding methods and finds that concept embedding achieves the highest performance. Additionally, the effectiveness of leveraging biomedical text embedding for drug safety prediction is demonstrated.
BMC BIOINFORMATICS
(2022)
Article
Genetics & Heredity
Chenfeng Wang, Hongdao Ma, Weiqing Wu, Xuhua Lu
Summary: This study identified core genes related to spinal cord injury (SCI) and ankylosing spondylitis (AS) through text mining and functional analysis. Potential drug targets were also identified. The findings provide valuable insights into the underlying mechanisms of these diseases and potential therapeutic approaches.
FRONTIERS IN GENETICS
(2022)
Article
Biochemical Research Methods
Zhi-Hui Luo, Li-Da Zhu, Ya-Min Wang, Sheng Hu Qian, Menglu Li, Wen Zhang, Zhen-Xia Chen
Summary: Disease pathogenesis is a significant topic in biomedical research. Through text mining, a new disease research tool called DSEATM has been developed and successfully applied to 3250 diseases, providing fresh insights into disease-related pathways.
BRIEFINGS IN BIOINFORMATICS
(2022)
Review
Information Science & Library Science
Sudha Cheerkoot-Jalim, Kavi Kumar Khedo
Summary: This work presents the results of a systematic literature review on biomedical text mining, identifying different text mining approaches, common tools, and challenges in the field. Researchers primarily utilize data sources such as electronic health records, biomedical literature, social media, and health-related forums for text mining. The most common technique is natural language processing, often using tools like MetaMap and Unstructured Information Management Architecture.
JOURNAL OF KNOWLEDGE MANAGEMENT
(2021)
Article
Computer Science, Artificial Intelligence
Matthew Shardlow, Luciano Gerber, Raheel Nawaz
Summary: This work uncovers the polysemous nature of emoji and develops a corpus to predict their meanings, highlighting the importance of considering the meaning behind emoji.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Biochemical Research Methods
Eryk Kropiwnicki, Alexander Lachmann, Daniel J. B. Clarke, Zhuorui Xie, Kathleen M. Jagodnik, Avi Ma'ayan
Summary: DrugShot is a web-based server application that provides ranked lists of drugs and small molecules relevant to biomedical search terms. By integrating PubMed abstracts and LINCS L1000 signatures, DrugShot can predict additional drugs and small molecules associated with the search term.
BMC BIOINFORMATICS
(2022)
Article
Pharmacology & Pharmacy
Kevin McCoy, Sateesh Gudapati, Lawrence He, Elaina Horlander, David Kartchner, Soham Kulkarni, Nidhi Mehra, Jayant Prakash, Helena Thenot, Sri Vivek Vanga, Abigail Wagner, Brandon White, Cassie S. Mitchell
Summary: Link prediction in artificial intelligence is utilized for identifying missing links or predicting future relationships in complex networks. A model using a complex heterogeneous biomedical knowledge graph called SemNet was developed to predict missing links in biomedical literature for drug discovery. The model achieved high accuracy in entity prediction tasks and was demonstrated through a case study on COVID-19 for drug discovery purposes.
Article
Computer Science, Information Systems
Muhammad Azeem Sarwar, Mansoor Ahmed, Asad Habib, Muhammad Khalid, M. Akhtar Ali, Mohsin Raza, Shahid Hussain, Ghufran Ahmed
Summary: The proposed framework introduces an ontology recommendation system using text categorization and unsupervised learning techniques to address the challenges of selecting appropriate ontologies. It organizes ontologies based on domain expert opinions and recommends them according to user requirements.
Article
Chemistry, Multidisciplinary
Waseemullah, Zainab Fatima, Shehnila Zardari, Muhammad Fahim, Maria Andleeb Siddiqui, Ag. Asri Ag. Ibrahim, Kashif Nisar, Laviza Falak Naz
Summary: Text summarization is a technique for shortening long texts or documents. Manual summarization can be costly and time-consuming, while an extractive summarization model balances compression and retention ratios by preserving meaningful sentences and filtering out redundant information.
APPLIED SCIENCES-BASEL
(2022)
Article
Pharmacology & Pharmacy
Jiang-Shan Tan, Song Hu, Ting-Ting Guo, Lu Hua, Xiao-Jian Wang
Summary: By text mining and functional enrichment analysis, we identified potential drugs that may have therapeutic effects on connective tissue disease-associated pulmonary arterial hypertension (CTD-PAH).
FRONTIERS IN PHARMACOLOGY
(2022)
Article
Biology
Khorshed Alam, Md Mahmudul Islam, Kai Gong, Muhammad Nazeer Abbasi, Ruijuan Li, Youming Zhang, Aiying Li
Summary: As an emerging resource, Gram-negative Burkholderia bacteria have the potential to produce a wide range of bioactive secondary metabolites. Genome mining has revealed significant diversity in biosynthetic gene clusters (BGCs) among different Burkholderia species, suggesting a high potential for biotechnological applications. Further investigation of BGC distributions within Burkholderia species may lead to the discovery of novel drug molecules for potential therapeutic use.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Genetics & Heredity
Hui-O Chen, Peng-Chan Lin, Chen-Ruei Liu, Chi-Shiang Wang, Jung-Hsien Chiang
Summary: Developing a biomedical text mining pipeline can help in cancer gene panel discovery. The researchers built a gene term-feature matrix using text-mined co-occurrence features and validated the panel's accuracy. They demonstrated the predictive power of using Text Mining technology for cancer gene prediction.
FRONTIERS IN GENETICS
(2021)
Article
Computer Science, Information Systems
Ying Hu, Yanping Chen, Ruizhang Huang, Yongbin Qin, Qinghua Zheng
Summary: Biomedical relation extraction aims to extract the interactive relations between biomedical entities in a sentence. This study proposes a hierarchical convolutional model to address the semantic overlapping and data imbalance problems. The model encodes both local contextual features and global semantic dependencies, enhancing the discriminability of the neural network for biomedical relation extraction.
INFORMATION PROCESSING & MANAGEMENT
(2024)
Article
Clinical Neurology
Chenfeng Wang, Liang Wang, Qisheng Li, Weiqing Wu, Jincan Yuan, Haibin Wang, Xuhua Lu
Summary: By using text mining and pathway analysis, this study identified genes and biological pathways associated with ankylosing spondylitis (AS) and osteoporosis (OP), providing potential drug targets for the prevention and treatment of AS-induced OP.
WORLD NEUROSURGERY
(2023)
Article
Computer Science, Hardware & Architecture
Ashutosh Kumar, Aakanksha Sharaff
Summary: SnorkelPlus model is proposed to extract biomedical relations between gene and disease entities from unstructured biomedical text without human effort. It achieves an AUROC of 85.60% and an AUPR of 45.73%, outperforming the baseline model, and creates a gene-disease relation database from 29 million scientific abstracts.