Related Experiment Video
Updated: Sep 11, 2025
08:51
Author Spotlight: Integrated Multi-Omics Analysis for Unveiling Multicellular Immune Signatures in Clinical Heart Attack Cohorts
Published on: September 20, 2024
1.5K
Systemic Lupus Erythematosus prediction using Epistatic-Quantile Fusion Transformer network with integrated
Manoj B Chandak1, Abhijeet R Raipurkar1, Sunita G Rawat1
1Ramdeobaba University, Katol Road, Nagpur-440013, India.
Computational Biology and Chemistry
|August 16, 2025
Summary
A new Epistatic-Quantile Fusion Transformer (EQF-T) model accurately predicts Systemic Lupus Erythematosus (SLE) by integrating multi-omics and EHR data. This advanced framework enhances early diagnosis of the complex autoimmune disorder.
Area of Science:
- Computational biology
- Autoimmune disease research
- Machine learning in healthcare
Background:
- Systemic Lupus Erythematosus (SLE) presents diagnostic challenges due to heterogeneous symptoms and overlapping clinical features.
- Traditional predictive models struggle with high-dimensional multi-omics data, biological variability, and data imbalance inherent in SLE research.
Purpose of the Study:
- To introduce a unified framework, the Epistatic-Quantile Fusion Transformer (EQF-T), for effective integration and analysis of multi-omics and Electronic Health Records (EHR) data.
- To improve the accuracy and reliability of early Systemic Lupus Erythematosus (SLE) prediction.
Main Methods:
- Utilized Beta-Variational Rank-ordered Quantile Autoencoder (Beta-VARQA) for denoising and normalizing heterogeneous biological data.
- Employed Epistatic Attention fused Multi-Omics Laplacian Transformer (EA-MLT) to capture gene-gene interactions and structural dependencies across omics layers.
- Developed SLE-Net, an end-to-end deep learning model for classification and interpretable output generation.
Main Results:
- The EQF-T framework achieved exceptional performance metrics: 99.82% accuracy, 99.78% precision, 99.76% recall, 99.77% F1-score, and 99.8% ROC-AUC.
- Demonstrated the model's capability to effectively learn from complex, high-dimensional biological and clinical data for precise SLE prediction.
Conclusions:
- The proposed EQF-T model offers a reliable and highly accurate approach for the early prediction of Systemic Lupus Erythematosus (SLE).
- The framework's novel components effectively address limitations in integrating diverse biological and clinical datasets for complex disease modeling.
