Predictive evolutionary genomics: principles, validation, and practice

Area of Science:

• Evolutionary biology
• Conservation science
• Agricultural science

Background:

• Climate change and habitat loss are accelerating evolutionary changes globally.
• Existing evolutionary forecasting methods operate on different timescales and remain disconnected.
• There is an urgent need for integrated approaches to predict evolutionary responses.

Purpose of the Study:

• To present a unified Bayesian framework for evolutionary prediction.
• To integrate trait-based, allele-based, and composite adaptation score approaches.
• To provide probabilistic forecasts with explicit uncertainty.

Main Methods:

• Developed a Bayesian framework integrating genomic, phenotypic, and environmental data.
• Combined three complementary evolutionary forecasting approaches: trait-based, allele-based, and composite adaptation scores.
• Validated forecasts using experimental evolution, field experiments, historical specimens, and reciprocal transplants.

Main Results:

• The integrated framework yields probabilistic evolutionary predictions.
• Forecasts can be validated using diverse empirical methods.
• The approach successfully combines predictions across multiple time scales.

Conclusions:

• This integrated framework advances evolutionary forecasting for conservation and agriculture.
• It enables prediction of extinction risks, optimization of breeding programs, and ecosystem management.
• Supports a shift towards predictive evolutionary biology for biodiversity and food security.