Genome & Epigenome
Genetic Scoring
Discover genetic risk scoring and HeraBiyo's unique methodology.
Genetic Risk Scoring
Polygenic Risk Score (PRS) is an advanced bioinformatics method that estimates an individual's genetic susceptibility to complex diseases by aggregating the effects of thousands or even millions of genes through gene–environment interactions.
With this method, you can learn your genetic risk for multifactorial diseases such as neurological disorders, cardiovascular diseases, diabetes, and cancer as a numerical score.
HeraBiyo's Unique PRS Methodology
To reduce predictive power limitations across diverse genetic backgrounds and improve clinical accuracy, HeraBiyo develops PRS models calibrated with local population data and validated with clinical endpoints.
HeraBiyo delivers tailor-made reports customized for genetic diagnostics workflows, ensuring seamless adoption and high customer satisfaction.
Advanced Genotype Processing & Standardization
Depending on the data source (WGS or SNP Array), raw data is genotyped with BWA/GATK or processed via CEL/IDAT with calling and imputation (Eagle v2/SHAPEIT4). Using PLINK, data is formatted for scoring and QC steps are completed.
Local Population Calibration
Variant weights from international studies are applied based on an individual's genetic ancestry to mitigate performance drop across populations and enhance predictive power for local cohorts.
Integration with Machine Learning
PRS and ancestry are combined with core clinical attributes (age, gender, blood pressure, diabetes, lipids, smoking, etc.) via multi-layer perceptron models to improve risk prediction accuracy.
References
- Steinfeldt, J. et al. (2022). Neural network-based integration of polygenic and clinical information... The Lancet Digital Health, 4(2).
- Lennon, N. J. et al. (2023). Selection, optimization, and validation of ten chronic disease polygenic risk scores... medRxiv.
- Wang, Y. et al. (2022). Challenges and Opportunities for Developing More Generalizable Polygenic Risk Scores. Annual review of biomedical data science, 5.
- Ge, T. et al. (2019). Polygenic prediction via Bayesian regression and continuous shrinkage priors. Nat Commun 10.