SynDx
Syndemic Vulnerability Engine (SVE)
Submitted & Under Review
Status
Validation manuscript submitted. Deployment manuscript in preparation.
Project By:
Project Description
PhenoGenX is a dual-engine computational platform designed to enhance HIV-1 drug resistance interpretation by integrating rule-based mutation frameworks with ensemble machine learning–driven phenotypic prediction. The system leverages curated resistance mutation databases alongside data-driven modeling to provide robust, scalable, and clinically relevant insights into antiretroviral susceptibility.
Status: Manuscript under review; preprint available at: https://www.researchsquare.com/article/rs-9056343/v1
Approach
The platform combines two complementary analytical engines:
- Rule-based interpretation, grounded in established HIV drug resistance mutation algorithms and expert-curated knowledge bases.
- Ensemble machine learning models, trained on large-scale genotype–phenotype datasets to predict drug susceptibility with improved accuracy and generalizability.
These components are integrated within a unified framework that enables cross-validation between mechanistic and data-driven predictions, improving interpretability while maintaining predictive performance. The system is designed to support modular expansion, allowing incorporation of new datasets, mutation rules, and therapeutic agents as they emerge.
Key Finding
Key Findings
- The dual-engine approach improves consistency and robustness of resistance interpretation compared to single-method frameworks.
- Ensemble machine learning models demonstrate strong predictive performance across diverse mutation profiles.
- Integration of rule-based and data-driven methods enhances interpretability while preserving scalability.
- The platform shows potential for clinical decision support in resource-constrained and high-burden settings.
