• English
  • Deutsch
  • Log In
    Password Login
    Research Outputs
    Fundings & Projects
    Researchers
    Institutes
    Statistics
Repository logo
Fraunhofer-Gesellschaft
  1. Home
  2. Fraunhofer-Gesellschaft
  3. Konferenzschrift
  4. ProtSTonKGs: A Sophisticated Transformer Trained on Protein Sequences, Text, and Knowledge Graphs
 
  • Details
  • Full
Options
January 10, 2022
Conference Paper
Title

ProtSTonKGs: A Sophisticated Transformer Trained on Protein Sequences, Text, and Knowledge Graphs

Abstract
While most approaches individually exploit unstructured data from the biomedical literature or structured data from biomedical knowledge graphs, their union can better exploit the advantages of such approaches, ultimately improving representations of biology. Using multimodal transformers for such purposes can improve performance on context dependent classification tasks, as demonstrated by our previous model, the Sophisticated Transformer Trained on Biomedical Text and Knowledge Graphs (STonKGs). In this work, we introduce ProtSTonKGs, a transformer aimed at learning all-encompassing representations of protein-protein interactions. ProtSTonKGs presents an extension to our previous work by adding textual protein descriptions and amino acid sequences (i.e., structural information) to the text- and knowledge graph-based input sequence used in STonKGs. We benchmark ProtSTonKGs against STonKGs, resulting in improved F1 scores by up to 0.066 (i.e., from 0.204 to 0.270) in several tasks such as predicting protein interactions in several contexts. Our work demonstrates how multimodal transformers can be used to integrate heterogeneous sources of information, paving the foundation for future approaches that use multiple modalities for biomedical applications.
Author(s)
Balabin, Helena
Fraunhofer-Institut für Algorithmen und Wissenschaftliches Rechnen SCAI  
Hoyt, Charles Tapley
Laboratory of Systems Pharmacology, Harvard Medical School
Gyori, Benjamin
Laboratory of Systems Pharmacology, Harvard Medical School
Bachman, John
Laboratory of Systems Pharmacology, Harvard Medical School
Tom Kodamullil, Alpha
Fraunhofer-Institut für Algorithmen und Wissenschaftliches Rechnen SCAI  
Hofmann-Apitius, Martin  
Fraunhofer-Institut für Algorithmen und Wissenschaftliches Rechnen SCAI  
Domingo Fernández, Daniel  
Fraunhofer-Institut für Algorithmen und Wissenschaftliches Rechnen SCAI  
Mainwork
SWAT4HCLS 2022, Semantic Web Applications and Tools for Health Care and Life Sciences  
Conference
International Conference on Semantic Web Applications and Tools for Health Care and Life Sciences 2022  
Open Access
File(s)
Download (1.83 MB)
Rights
CC BY
DOI
10.24406/publica-72
Language
English
Fraunhofer-Institut für Algorithmen und Wissenschaftliches Rechnen SCAI  
Keyword(s)
  • Natural Language Processing

  • Knowledge Graphs

  • Transformers

  • Bioinformatics

  • Machine Learning

  • Cookie settings
  • Imprint
  • Privacy policy
  • Api
  • Contact
© 2024