Related Experiment Video
Updated: Jun 25, 2025
10:44
Sex Stratified Neuronal Cultures to Study Ischemic Cell Death Pathways
Published on: December 9, 2013
12.0K
SPIN: sex-specific and pathway-based interpretable neural network for sexual dimorphism analysis
Euiseong Ko1, Youngsoon Kim2, Farhad Shokoohi3
1Department of Computer Science, University of Nevada, Las Vegas, Las Vegas, NV, USA.
Briefings in Bioinformatics
|May 29, 2024
Summary
This study introduces SPIN, a deep learning framework for analyzing sex differences in diseases. SPIN improves risk prediction and identifies sex-specific genetic factors, paving the way for personalized medicine.
Area of Science:
- Genetics
- Computational Biology
- Precision Medicine
Background:
- Most common diseases exhibit sexual dimorphism in prevalence, severity, and genetic susceptibility.
- Current genetic and clinical studies often overlook sex-specific effects by analyzing data in a combined framework or separately, failing to detect gene-by-sex interactions.
Purpose of the Study:
- To propose a novel, unified, and biologically interpretable deep learning framework (SPIN) for analyzing sexual dimorphism in disease.
- To improve risk prediction and identify sex-specific genetic risk loci.
Main Methods:
- Development of a deep learning framework named SPIN (Sex-specific Prediction and Interpretation Network).
- Application and validation of SPIN on TCGA cancer datasets and asthma datasets.
Main Results:
- SPIN significantly improved the C-index (a measure of prediction accuracy) by up to 23.6% in TCGA cancer datasets.
- SPIN successfully identified sex-specific and shared risk loci missed by conventional methods.
- The framework demonstrated interpretability in explaining biological pathway contributions to sexual dimorphism and individual risk.
Conclusions:
- SPIN offers a powerful, interpretable approach for analyzing sexual dimorphism in disease genetics and clinical outcomes.
- This framework enhances disease risk prediction and facilitates the discovery of novel sex-specific genetic factors.
- SPIN holds potential for developing precision medicine strategies tailored to individual characteristics, considering sex as a key biological factor.
