Evolution is more repeatable in the introduction than range expansion phase of colonization

How repeatable is evolution at genomic and phenotypic scales? We studied the repeatability of evolution during 8 generations of colonization using replicated microcosm experiments with the red flour beetle, Tribolium castaneum. Based on the patterns of shared allele frequency changes that occurred in populations from the same generation or experimental location, we found adaptive evolution to be more repeatable in the introduction and establishment phases of colonization than in the spread phase, when populations expand their range. Lastly, by studying changes in allele frequencies at conserved loci, we found evidence for the theoretical prediction that range expansion reduces the efficiency of selection to purge deleterious alleles. Overall, our results increase our understanding of adaptive evolution during colonization, demonstrating that evolution can be highly repeatable, while also showing that stochasticity still plays an important role., , , # Evolution is more repeatable in the introduction than range expansion phase of colonization

Code and data associated with analysis and figures.

## Sharing/Access information

The full list of NCBI Sequence Read Archive accessions can be found at

## Code/Software

Figures were generated with R, scripts split up by figure/analysis type.

### Details by file

#### Data

* BayPass_manual.pdf Baypass Software Manual (Version 2.1). Includes complete description of output files and the variables therein starting on page 21.

* Z.csv
* Key relating sequence library names to treatments. Columns are sample prefix and dummy variables to indicate which treatment each sample was from (founder, core, edge, shuffled).

* baypass_r0.98_d3_L3_M30_q0.99_a35.txt, tribolium.covariates, tribolium.poolsize
* Input files required to run Baypass and generate outlier loci. See BayPass_manual.pdf for more details about file formats.

* aux_model_summary_yij_pij.out, aux_model_summary_...