Table 4 Results of generalized linear mixed models (GLMM) testing the influence of geographic (GEO) distance and environmental distances, derived from climate (CLIM), soil (SOIL), and elevation (ELEV) data, on the genetic divergence among 16 populations of Pitcairnia lanuginosa from the Cerrado.

From: Evidence of local adaptation despite strong drift in a Neotropical patchily distributed bromeliad

Response variable

Outlier signal

Model

DIC

ΔDIC

DIC weight

Neutral SNPs

Null

338.245

4.667

0.041

GEO

337.829

4.251

0.051

CLIM

339.737

6.159

0.020

SOIL

333.578

0.000

0.425

ELEV

339.669

6.091

0.020

GEO + CLIM

339.238

5.660

0.025

GEO + SOIL

334.669

1.091

0.247

GEO + ELEV

339.714

6.136

0.020

CLIM + SOIL + ELEV

337.133

3.556

0.072

GEO + CLIM + SOIL + ELEV

336.928

3.351

0.080

Outlier SNP 1

Divergent selection

Null

156.188

17.873

0.000

GEO

143.053

4.739

0.045

CLIM

156.459

18.145

0.000

SOIL

150.881

12.567

0.001

ELEV

157.841

19.526

0.000

GEO + CLIM

139.240

0.926

0.301

GEO + SOIL

142.187

3.872

0.069

GEO + ELEV

141.325

3.011

0.106

CLIM + SOIL + ELEV

152.516

14.202

0.000

GEO + CLIM + SOIL + ELEV

138.315

0.000

0.478

Outlier SNP 2

Divergent selection

Null

156.194

17.881

0.000

GEO

143.046

4.733

0.045

CLIM

156.473

18.160

0.000

SOIL

150.873

12.560

0.001

ELEV

157.840

19.526

0.000

GEO + CLIM

139.249

0.936

0.300

GEO + SOIL

142.184

3.871

0.069

GEO + ELEV

141.326

3.013

0.106

CLIM + SOIL + ELEV

152.510

14.197

0.000

GEO + CLIM + SOIL + ELEV

138.313

0.000

0.479

  1. Analyses were performed using either an FST matrix derived from 616 neutral SNPs, or the allele frequency differential delta (AFD) for two candidate outlier SNPs. The best models for each data set are highlighted in bold (ΔDIC  <  2).