Work

Integration of literature and data for context-aware model curation: a glioblastoma stem cell case study

Year: 2023
Type: article
Abstract: Computational modeling serves many purposes in biomedical research. In addition to understanding mechanisms of normal healthy cell function, computational modeling also provides valuable insights into the mechanisms of disease. In recent years, automated tools for curating computational models of cell function have become more accurate and widespread. However, many obstacles remain for automated modeling in a personalized medicine context. First, many models of disease signaling are merely interaction networks, and do not encode information about rules for dynamic signaling behavior. Additionally, many of these models are not comprehensive enough to make widespread conclusions about the effect of disease control interventions. While automated information retrieval speeds up model curation, machine learning approaches for extracting signaling events from literature are not trustworthy enough to use without human intervention. This dissertation will attempt to address several of these obstacles through a glioblastoma multiforme (GBM) stem cell case study. By utilizing discrete modeling techniques, this GBM model is able to capture the progression of disease at multiple levels of specificity. To address the inaccuracies and natural language processing results I also present a tool for using database results to judge machine-reading. Altogether, the GBM case study and methodology presented in this dissertation can serve as a guide for personalized, automated modeling of disease.
Language: English
Open Access status: green
OpenAlex ID: W7066627072
FWCI:
Cites: 122
Cited by:
Related to: 10