Background: Biomedical event extraction is a fundamental and in-demand technology that has attracted substantial interest from many researchers. Previous works have heavily relied on manual designed features and external NLP packages in which the feature engineering is large and complex. Additionally, most of the existing works use the pipeline process that breaks down a task into simple sub-tasks but ignores the interaction between them. To overcome these limitations, we propose a novel event combination strategy based on hybrid deep neural networks to settle the task in a joint end-to-end manner. Results: We adapted our method to several annotated corpora of biomedical event extraction tasks. Our method achieved state-of-the-art performance with noticeable overall F1 score improvement compared to that of existing methods for all of these corpora. Conclusions: The experimental results demonstrated that our method is effective for biomedical event extraction. The combination strategy can reconstruct complex events from the output of deep neural networks, while the deep neural networks effectively capture the feature representation from the raw text. The biomedical event extraction implementation is available online at http://www.predictor.xin/event_extraction.
Altered metabolism is often identified as a cause or an effect of physiology and pathogenesis. But it is difficult to predict the metabolic flux distributions of multicellular organisms due to the lack of an explicit metabolic objective function. Here we present a computational method which can successfully describe the differences in metabolism between two different conditions on a large scale. By integrating gene expression data with an existing comprehensive reconstruction of the global human metabolic network, we qualitatively predicted significantly differential fluxes without prior knowledge or the rate of metabolite uptake and secretion. Therefore, this method can be applied for both microorganisms and multicellular organisms. Different from traditional enrichment analysis methods and constraint-based models, we consider conditions and interactions within the metabolic network simultaneously. To apply the proposed method, we predicted altered fluxes for E. coli strains and clear cell renal cell carcinoma, while the E. coli strains are growing aerobically in a chemostat with different dilution rates and clear cell renal cell carcinoma is compared with normal kidney cells. Then we map the significantly differential reactions to metabolic subsystems defined in the original metabolic network for ccRCC to observe the altered metabolism. In contrast with existing studies, our results show a high accuracy of the E. coli experiment and a more reasonable prediction of the ccRCC experiment. The method presented here provides a computational approach for the genome-wide study of altered metabolism under pairs of conditions for both microorganisms and multicellular organisms.
ABSTRACT. This study aims to identify the crucial miRNAs in Epstein-Barr virus-positive nasopharyngeal carcinoma (NPC) and their target genes. Gene expression profile data (GSE12452) that included 31 NPC and 10 normal nasopharyngeal tissue specimens were downloaded. Differentially expressed genes (DEGs) were identified using significance analysis of microarrays. The underlying function of DEGs was predicted via Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathway enrichment analyses. The miRNA sequencing dataset GSE14738 was also downloaded, and expression levels of miRNA were calculated by the number of reads mapped to each miRNA. The selected miRNAs were integrated into the miRecords 6029 Integrated miRNA-mRNA analysis of NPC ©FUNPEC-RP www.funpecrp.com.br Genetics and Molecular Research 14 (2): 6028-6036 (2015) database to obtain their target genes. Target genes associated with DEGs were used to construct the interaction network via Cytoscape. A total of 1437 DEGs between NPC and control were identified, most of which were enriched in cell cycle and extracellular matrix-receptor interaction signaling pathways. Furthermore, 112 miRNAs were considered upregulated in NPC samples. A total of 2228 relationships between 39 miRNAs and 1247 target genes were obtained, of which 182 relationships between 32 miRNAs and 97 target genes were chosen to construct an interaction network. The interactions between DEGs and the let-7 or miR-29 families appeared strongest in this network, where CDC25A, COL3A1, and COL1A1 were regulated by several let-7 family members, while COL4A1 and COL5A2 were regulated by several miR-29 family members. The let-7 and miR-29 families may be related to the development of NPC by regulating the genes involved in cell cycle and ECM-receptor interaction.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.