Related Experiment Video
Updated: Aug 12, 2025
14:27
Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
Published on: June 26, 2013
15.7K
SWEET: a single-sample network inference method for deciphering individual features in disease
Hsin-Hua Chen1, Chun-Wei Hsueh1, Chia-Hwa Lee2,3,4
1Institute of Bioinformatics and Systems Biology, National Yang Ming Chiao Tung University, Hsinchu 300, Taiwan.
Briefings in Bioinformatics
|January 31, 2023
Summary
We developed a new method called SWEET to build accurate single-sample networks (SINs) from gene expression data. This approach improves personalized cancer diagnosis and treatment by revealing individual patient characteristics and identifying potential drug targets.
Area of Science:
- Genomics
- Systems Biology
- Network Medicine
Background:
- Accurate construction of single-sample networks (SINs) is crucial for personalized diagnostics and therapeutics.
- Current methods struggle to capture individual patient characteristics and disease heterogeneity.
Purpose of the Study:
- To introduce the sample-specific-weighted correlation network (SWEET) method for improved SIN inference.
- To enhance the characterization of individual biological systems and disease subtypes.
Main Methods:
- Developed the SWEET method integrating genome-wide sample-to-sample correlation with differential network analysis.
- Assessed genome-wide sample weights without prior subpopulation knowledge to mitigate bias.
- Validated SWEET SINs against scale-free properties, human interactomes, and cancer gene identification.
Main Results:
- SWEET SINs demonstrated superior scale-free properties and overlap with human interactomes across 16 cancers.
- The method effectively identified cancer-related genes, somatic mutations, mut-drivers, and essential genes.
- Identified two candidate drugs (albendazole, encorafenib) and two LUAD subtypes with distinct features.
Conclusions:
- SWEET offers a robust approach for SIN inference, complementing existing methods.
- The method provides valuable insights for network medicine and precision medicine applications.
- SWEET facilitates the identification of novel therapeutic strategies and disease subtypes.
