Collaborative Proposal: Accelerating Synthetic Biology Discovery & Exploration through Knowledge Integration

合作提案：加速合成生物学发现

• 批准号: 1939951
• 负责人: Bridget McInnes
• 金额: $ 15.58万
• 依托单位: Virginia Commonwealth University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-10-01 至 2023-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1939951&HistoricalAwards=false
• 关键词: Collaborative; Proposal; Accelerating; Synthetic; Biology

The scientific challenge for this project is to accelerate discovery and exploration of the synthetic biology design space. In particular, many parts used in synthetic biology come from or are initially tested in a simple bacteria, E. coli, but many potential applications in energy, agriculture, materials, and health require either different bacteria or higher level organisms (yeast for example). Currently, researchers use a trial-and-error approach because they cannot find reliable information about prior experiments with a given part of interest. This process simply cannot scale. Therefore, to achieve scale, a wide range of data must be harnessed to allow confidence to be determined about the likelihood of success. The quantity of data and the exponential increase in the publications generated by this field is creating a tipping point, but this data is not readily accessible to practitioners. To address this challenge, our multidisciplinary team of biological engineers, machine learning experts, data scientists, library scientists, and social scientists will build a knowledge system integrating disparate data and publication repositories in order to deliver effective and efficient access to collectively available information; doing so will enable expedited, knowledge-based synthetic biology design research.This project will develop an open and integrated synthetic biology knowledge system (SBKS) that leverages existing data repositories and publications to create a single interface that transforms the way researchers access this information. Access to up-to-date information in multiple, heterogeneous sources will be provided via a federated approach. New methods based on machine learning will be developed to automatically generate ontology annotations in order to create connections between data in various repositories and information extracted from publications. Provenance for each entity in SBKS will be tracked, and it will be utilized by new methods that are developed to assess bias and assign confidence scores to knowledge returned for each entity. An intuitive, natural-language-based interface and visualization functionality will be implemented for users to easily access and explore SBKS contents. Additionally, as ethics is necessarily a part of synthetic biology research, data from text sources related to ethical concerns in synthetic biology will also be incorporated to inform researchers about ethical debates relevant to their search queries. Finally, to test the SBKS API, a new genetic design tool, Kimera, will be developed that leverages the knowledge in SBKS to produce better designs. The proposed SBKS will accelerate discovery and innovation by enabling researchers to learn from others' past experiences and to maximize the productivity of valuable experimental time on testing designs that have a higher likelihood of working when transformed to a new organism. This research thus provides the potential for transformative research outcomes in the field of synthetic biology by leveraging data science to improve the field's epistemic culture. For more information please see https://synbioks.github.io.This project is part of the National Science Foundation's Harnessing the Data Revolution (HDR) Big Idea activity, and is jointly supported by the HDR and the Division of Biological Infrastructure within the NSF Directorate of Directorate for Biological Sciences.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

该项目的科学挑战是加速合成生物学设计空间的发现和探索。 特别是，合成生物学中使用的许多部分来自或最初在一种简单的细菌E。大肠杆菌，但在能源，农业，材料和健康的许多潜在应用需要不同的细菌或更高水平的生物体（例如酵母）。目前，研究人员使用试错法，因为他们无法找到关于给定感兴趣部分的先前实验的可靠信息。这个过程根本无法扩展。因此，为了实现规模化，必须利用广泛的数据，以确定成功的可能性。这一领域产生的数据量和出版物的指数增长正在创造一个临界点，但从业人员并不容易获得这些数据。为了应对这一挑战，我们的生物工程师，机器学习专家，数据科学家，图书馆科学家和社会科学家的多学科团队将建立一个整合不同数据和出版物存储库的知识系统，以便有效和高效地访问集体可用的信息;这样做将能够加快，基于知识的合成生物学设计研究。本项目将开发一个开放的、集成的合成生物学知识系统（SBKS）利用现有的数据存储库和出版物来创建一个单一的界面，改变研究人员访问这些信息的方式。将通过联合方法提供对多个不同来源的最新信息的访问。将开发基于机器学习的新方法，以自动生成本体论注释，从而在各种储存库中的数据与从出版物中提取的信息之间建立联系。 将跟踪SBKS中每个实体的来源，并将通过开发的新方法来使用它，以评估偏倚并为每个实体返回的知识分配置信度得分。一个直观的，自然语言为基础的界面和可视化功能将实现用户轻松访问和探索SBKS的内容。 此外，由于伦理学是合成生物学研究的一部分，来自合成生物学伦理问题相关文本来源的数据也将被纳入，以告知研究人员与其搜索查询相关的伦理辩论。 最后，为了测试SBKS API，将开发一种新的遗传设计工具Kimera，该工具利用SBKS中的知识来产生更好的设计。 拟议的SBKS将加速发现和创新，使研究人员能够从他人过去的经验中学习，并最大限度地提高宝贵的实验时间在测试设计上的生产力，这些设计在转化为新生物体时具有更高的工作可能性。 因此，这项研究通过利用数据科学来改善该领域的认知文化，为合成生物学领域的变革性研究成果提供了潜力。欲了解更多信息，请参阅https://synbioks.github.io.This项目是美国国家科学基金会利用数据革命（HDR）大创意活动的一部分，并由HDR和NSF生物科学理事会生物基础设施部共同支持。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Effects of data and entity ablation on multitask learning models for biomedical entity recognition

数据和实体消融对生物医学实体识别多任务学习模型的影响

• DOI: 10.1016/j.jbi.2022.104062
• 发表时间: 2022
• 期刊: Journal of Biomedical Informatics
• 影响因子: 4.5
• 作者: Rodriguez, Nicholas E.;Nguyen, Mai;McInnes, Bridget T.
• 通讯作者: McInnes, Bridget T.

BioCreative VII-Track 1: A BERT-based System for Relation Extraction in Biomedical Text

• 发表时间: 2022
• 作者: D. Mahendran;Sudhanshu Ranjan;Jia-Hong Tang;Mai H Nguyen;Bridget T. McInnes

An overview of biomedical entity linking throughout the years

• DOI: 10.1016/j.jbi.2022.104252
• 发表时间: 2022-12-09
• 期刊: JOURNAL OF BIOMEDICAL INFORMATICS
• 影响因子: 4.5
• 作者: French，Evan;McInnes，Bridget T.
• 通讯作者: McInnes，Bridget T.

Synthetic Biology Knowledge System

• DOI: 10.1021/acssynbio.1c00188
• 发表时间: 2021-08-13
• 期刊: ACS SYNTHETIC BIOLOGY
• 影响因子: 4.7
• 作者: Mante, Jeanet;Hao, Yikai;Myers, Chris J.
• 通讯作者: Myers, Chris J.