NALgen

NALgen (Neutral and Adaptive Landscape Genetics) is a method for modeling

the process of gene flow and predicting the resulting pattern of (neutral and

adaptive) genetic variation across (continuous) landscapes.

NALgen:

Analysis Steps

 To evaluate the NALgen method of modeling neutral and adaptive genetic variation across

continuous landscapes:

 Generated several landscapes, and then simulated genotypes based on the generated landscapes

 Used multivariate analyses

 to infer population structure in the simulated data

 to find covariance between genetic and environmental data.

 Modeled neutral and adaptive genetic variation across continuous landscapes

 Information about population structure used to create response variable

 Transformed environmental data (based on genetic-environmental covariance) used as

predictors

 Evaluated performance of model (using subsets of 20 randomly-distributed populations)

 Compared against original landscape used to simulate genotypes

 Compared against inferred map of ancestry based on genotypes simulated for each cell

(all 884 populations)

Landscape

Genetic

Simulations

 Generating simple (low-complexity) landscapes and simulating genotypes

 landscapeRpackage: simulating simple landscapes (barrier and two

different values of carrying capacity/habitat suitability): landscape 1 has

low permeability cells (carrying capacity = 5) between habitable patches

(carrying capacity = 100), while landscape 2 has high permeability cells

(carrying capacity = 25), in addition to barriers (carrying capacity = 0)

 popRangepackage: simulating 50selectedand 500neutralSNPs based

on thelandscapeR-generated landscape.

 Generating high-complexity landscapes and simulating genotypes

 virtualspeciespackage: landscape generation based on Bioclim data

 Collinearity removed from climatic variables,

 PCA of remaining variables performed,

 Ranges of PCA values specified (for three metapopulations).

 popRangepackage: simulating 50selectedand 500neutralSNPs based on

thevirtualspecies-generated landscape.

Simulated genotypes (50 selected and 500 neutral biallelic loci (SNPs), separately and combined): two migration coefficients

(m=0.1 and 0.5) x two levels of landscape complexity (low and high) x two levels of landscape permeability (low and high)

Low-Complexity

Low-Permeability

(LCLP) landscape

Low-Complexity

High-Permeability

(LCHP) landscape

High-Complexity

Low-Permeability

(HCLP) landscape

High-Complexity

High-Permeability

(HCHP) landscape

Setup of Evaluation Step:

884-population reference

The original landscapes used to

simulate the genetic data, as well

as population structure results

obtained from these complete

genetic datasets, will be used in

evaluating the NALgen method.

Presented hereinafter are the

reference results (mapped

ancestry coefficients) that NALgen

modeling based on 20 randomly-

distributed populations will be

compared against.

NALgen Method

To determine the maximum genetic-

environmental covariance, the raw data is

first transformed: these transformed

environmental scores form the predictors

in NALgen modeling

The response variables are measures of

genetic variation (neutral and adaptive):

these are newly-developed metrics