EnzChemRED,一个丰富的酶化学关系提取数据集
Po-Ting Lai1, Elisabeth Coudert2, Lucila Aimo2
1National Center for Biotechnology Information (NCBI), National Library of Medicine (NLM), National Institutes of Health (NIH), Bethesda, MD, 20894, USA.
酶修复被EnzChemRED加速,这是用于自然语言处理 (NLP) 模型的新数据集. 这使得可以从科学文献中自动提取酶功能,从而提高知识库的准确性.
科学领域:
- 生物化学 生物化学
- 生物信息学是一种生物信息学.
- 计算生物学 计算生物学
背景情况:
- 从科学文献中对酶功能的专家策划对于FAIR开放知识库至关重要.
- 目前的手工策划不能跟上新发现和出版的步伐.
- 需要自动化方法来协助和扩大酶修复工作.
研究的目的:
- 引入EnzChemRED (酶化学关系提取数据集),这是一个用于训练和比较自然语言处理 (NLP) 模型的新型数据集.
- 促进NLP方法的开发,包括大型语言模型,以协助酶修复.
- 能够从科学文献中可扩展地提取酶催化反应信息.
主要方法:
- EnzChemRED包括1,210个由专家策划的PubMed摘要.
- 酶和催化化学反应使用UniProtKB和ChEBI标识符进行注释.
- 语言模型使用EnzChemRED进行了微调,用于实体识别和关系提取.
主要成果:
- 精心调整的语言模型获得了高F1分:蛋白质和化学识别的86.30%.
- 化学转换的提取达到了86.66%的F1得分.
- 催化酶的提取实现了83.79%的F1得分.
结论:
- EnzChemRED显著提高了对酶功能提取的NLP模型性能.
- 开发的方法可以处理PubMed规模的摘要,以创建酶功能的草稿地图.
- 该资源将指导UniProtKB和Rhea的策划工作,改善生物知识库.
更多相关视频
13:00Hot Biological Catalysis: Isothermal Titration Calorimetry to Characterize Enzymatic Reactions
Published on: April 4, 2014
07:15Mechanism of Kemeng Fang's Inhibition of Podocyte Apoptosis in Rats with Membranous Nephropathy through the PI3K/AKT Signaling Pathway
Published on: August 23, 2024
相关概念视频
Introduction to Enzymes
Most enzymes are proteins that speed up biochemical reactions without being consumed. Enzymes contain one or more active sites that...
Catalytically Perfect Enzymes
Most enzymes...
Introduction to Enzyme Kinetics
The experimenter can then plot the initial reaction rate or velocity (Vo) of a given trial against the substrate concentration ([S]) to obtain a graph of the reaction properties. For many enzymatic reactions involving a...
Oxymercuration-Reduction of Alkenes
Restriction Enzymes
The host bacteria protect their own genomic DNA from these enzymes by methylating these sites. Some...
Enzyme-Linked Immunosorbent Assay
There are many different types of ELISAs, but they all involve an antibody molecule whose constant region binds an enzyme, leaving the variable region free to bind its specific antigen. Enzyme-substrate reaction allows the antigen to be visualized or...
