Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form
D-Scholarship @ Pitt Banner and Links to Homepage

Automated Translation Framework for Biological Event Annotation

Tang, Difei (2023) Automated Translation Framework for Biological Event Annotation. Master's Thesis, University of Pittsburgh. (Unpublished)

[img] PDF
Restricted to University of Pittsburgh users only until 13 June 2024.
Download (5MB) | Request a Copy

Abstract

Computational modeling improves our understanding of the components and dynamics of biological systems. However, current biological models are restricted in size and scope due to limitations resulting from manual curation and validation. While natural language processing (NLP) techniques now have the ability to capture detailed semantics by event extraction from large amounts of rapidly processed text, a gap still exists between these NLP event representations and modeling formalisms. If the events contained in the natural language of published biomedical literature, related scientific articles, could be extracted and translated accurately into generic representation of biological knowledge, the impact on computational modeling and analysis of complex biological systems would be significant.
In this thesis, we develop a standardized framework for translating the events found in semantic NLP event annotations. To capture complex event structures, especially nested events in which one event causes another, we first present an intermediate graph representation. Our framework then enables the extraction of causal interactions between biological entities by defining a set of translation rules. We also develop a SBML-compatible network format for creating reaction models from events. With this network, we extend the existing methods to facilitate the conversion from events to SBML reaction model. Here, we briefly introduce our approach and investigate how well event annotations can be translated into other representation formats without incorporating information from any external resources. We demonstrate the effectiveness of our framework by the automatic translation of selected event annotations. By standardizing the translation of events from NLP extractions, we propose this as a generalizable, scalable method for rapid, large-scale integration of knowledge on biological events extracted from literature.

Share

Citation/Export:
Social Networking:
Share |

Details

Item Type: University of Pittsburgh ETD
Status: Unpublished
Creators/Authors:
CreatorsEmailPitt UsernameORCID
Tang, DifeiDIT18@pitt.edu
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairMiskov-Zivanov, Natasanmzivanov@pitt.edu
Committee MemberAkcakaya, Muratakcakaya@pitt.edu
Committee MemberMao, Zhi-Hongzhm4@pitt.edu
Date: 13 June 2023
Date Type: Publication
Defense Date: 12 April 2023
Approval Date: 13 June 2023
Submission Date: 4 April 2023
Access Restriction: 1 year -- Restrict access to University of Pittsburgh for a period of 1 year.
Number of Pages: 59
Institution: University of Pittsburgh
Schools and Programs: Swanson School of Engineering > Electrical and Computer Engineering
Degree: MS - Master of Science
Thesis Type: Master's Thesis
Refereed: Yes
Uncontrolled Keywords: Event Extraction, knowledge representation, biological modeling
Date Deposited: 13 Jun 2023 14:21
Last Modified: 13 Jun 2023 14:21
URI: http://d-scholarship.pitt.edu/id/eprint/44418

Metrics

Monthly Views for the past 3 years

Plum Analytics

Actions (login required)

View Item View Item