The cutting-edge science behind Salud.ia

Yes. The platform supports polygenic analysis across different populations and genetic testing platforms. Models are built to account for differences in SNP coverage, reference populations, and data quality.

When large population datasets are available (such as UK Biobank), models are trained directly on those datasets. When that type of data is not available, models are built using findings from multiple peer-reviewed studies.

Each model clearly defines the population it was developed for and the level of interpretation it supports, rather than applying a one-size-fits-all approach.

Genetic risk scores can perform differently across ancestry groups. To address this, the platform adjusts how genetic variants are weighted based on population data and the strength of available scientific evidence.Variants are selected based on factors such as:

If a variant has strong scientific support but limited validation in a specific population, its influence is reduced and the confidence level of the insight is adjusted accordingly.

This helps ensure insights remain scientifically responsible while still using meaningful biological signals.

Two types of models are used depending on the available data.

These are trained directly on large population datasets (such as UK Biobank) where genetic and health outcome data are available. These models support more quantitative interpretation and undergo statistical validation.

When large datasets are not available, models are built from findings across published genetic studies. These models provide directional insights rather than precise risk estimates.

Once released, each model stays the same to ensure consistent and reproducible results until new versions are thoroughly researched and established by our science team

The platform interprets health data using multiple inputs rather than relying on genetics alone.

Different data sources provide different types of insight:

Insights come from evaluating how these signals work together rather than relying on a single metric.

La plataforma no presenta puntajes genéticos crudos ni valores de laboratorio por sí mismos. En cambio, los datos biológicos se interpretan a través de modelos estructurados que traducen la ciencia compleja en ideas claras. Los resultados se evalúan en contexto y se limitan a lo que la evidencia científica puede respaldar. Las ideas ayudan a explicar cómo pueden funcionar diferentes sistemas biológicos para un individuo y por qué ciertas estrategias de nutrición, recuperación o entrenamiento pueden ser más adecuadas. A medida que se agregan datos adicionales, como laboratorios, datos de dispositivos portátiles o entradas de estilo de vida, las recomendaciones pueden volverse más específicas mientras se mantiene la interpretación genética subyacente consistente.

Insights are prioritized based on several factors, including:

This helps ensure the most reliable signals are highlighted first.

The platform improves through model updates and new scientific evidence, not by changing results in real time.

When models are updated or improved, new versions are released. Each version remains fixed for users evaluated under that model to ensure consistency and transparency.