Related Experiment Video
Updated: May 30, 2026

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
Published on: October 11, 2018
Ranking support vector machine for multiple kernels output combination in protein-protein interaction extraction from
Zhihao Yang1, Yuan Lin, Jiajin Wu
1College of Computer Science and Technology, Dalian University of Technology, Dalian, P. R. China. yangzh@dlut.edu.cn
This study introduces a novel multiple kernel learning approach for automatically extracting protein-protein interactions (PPIs) from biomedical literature. The method significantly improves the accuracy of identifying these crucial biological interactions.
Area of Science:
- Biomedical Informatics
- Computational Biology
- Bioinformatics
Background:
- Understanding protein-protein interactions (PPIs) is vital for elucidating biological mechanisms.
- The rapid growth of biomedical literature poses challenges for manual curation of PPI databases.
- Automated methods are needed to efficiently extract PPI information.
Purpose of the Study:
- To develop and evaluate an automated approach for extracting protein-protein interactions from biomedical texts.
- To improve the efficiency and accuracy of PPI database curation.
Main Methods:
- A multiple kernel learning framework was employed for PPI extraction.
- The approach integrated feature-based, tree, and graph kernels.
- A Ranking Support Vector Machine (SVM) was used to combine kernel outputs.
Main Results:
- Individual kernels demonstrated complementary features for PPI extraction.
- The combined kernel approach with Ranking SVM outperformed individual kernels and simple weight combinations.
- The method achieved state-of-the-art performance with a 64.88% F-score and 88.02% AUC on the AImed corpus.
Conclusions:
- The proposed multiple kernel learning approach effectively automates protein-protein interaction extraction from literature.
- This method offers a significant advancement for curating large-scale PPI databases.
- The findings highlight the utility of integrating diverse features for enhanced biological information extraction.
Related Concept Videos
Protein-protein Interfaces
Protein-Protein Interfaces
Protein Networks
These interactions can be represented through maps depicting protein-protein interaction networks, represented as nodes and edges. Nodes are circles that are representative of a protein,...
Protein Networks
These interactions can be represented through maps depicting protein-protein interaction networks, represented as nodes and edges. Nodes are circles that are representative of a protein,...
Conserved Binding Sites
Binding sites are often located in large pockets, and if their location on a protein’s surface is unknown, it can be predicted using various approaches. The energetic method computationally analyses the...
Protein Complexes with Interchangeable Parts
The SCF ubiquitin ligase is a protein complex of five individual proteins. This complex attaches ubiquitin to other target proteins to mark them for degradation. In order to...

