AlleleShift

AlleleShift predicts population-level changes in allele frequencies in response to climate change using R-based analyses for evolutionary and conservation genetics.

Key Features:

• Predictive Methodology: Uses a two-step predictive workflow combining redundancy analysis (RDA) and generalized additive models (GAM) to estimate climate-driven allele frequency changes.
• Redundancy Analysis (RDA): Performs RDA using Euclidean distances analogous to those used in Analysis of Molecular Variance (AMOVA) to establish baseline allele–environment relationships.
• Generalized Additive Model (GAM): Calibrates RDA-derived predictions with a GAM using a binomial family to constrain predicted allele frequencies to the 0–1 interval.
• Visualization Capabilities: Generates dot plots (shift.dot.ggplot), pie diagrams (shift.pie.ggplot), moon diagrams (shift.moon.ggplot), waffle diagrams (shift.waffle.ggplot), and smoothed geographic surface diagrams (shift.surf.ggplot).
• Animation and Temporal Analysis: Supports creation of animations to illustrate temporal changes in allele frequencies using ggplot2.
• Data Requirements: Accepts genetic input formatted as adegenet::genpop (derivable from adegenet::genind) and climatic layers from sources such as WorldClim and Envirem.

Scientific Applications:

• Local Adaptation Studies: Predicts how allele frequencies associated with adaptive traits may shift under future or historical climate scenarios to inform evolutionary inference.
• Conservation Planning: Identifies populations at risk from climate-driven genetic change to inform assisted migration and other conservation strategies.
• Paleoecological Research: Enables exploration of allele frequency changes across historical and future climates to provide broader evolutionary context.

Methodology:

Performs redundancy analysis (RDA) using Euclidean distances akin to AMOVA to model allele–environment relationships, then fits a generalized additive model (GAM) with a binomial family to calibrate predictions within the 0–1 range.