## Similar protocols

## Protocol publication

[…] We sourced published population-level DNA sequence data from Genbank (http://www.ncbi.nlm.nih.gov/genbank/) for broadly distributed Indo-Pacific reef-associated fish and invertebrates with a partial distribution in the Coral Triangle region. We used data for species that had sequences from three or more locations, where eight or more individuals had been sampled from each, and where at least one of the locations sampled was within 2500 km of our approximated centre of the Coral Triangle (approximated as 1°35′ S, 135°20′ E, following maps in Green & Mous []; ). In total, this yielded data from 45 species, 314 populations in total and an average of 6.97 (range 3 to 23) populations per species (electronic supplementary material, table S1). We aligned data for each species separately using ClustalW in **DAMBE** v. 5.3 [], and reconstructed the effective population size through time for each population separately using Bayesian skyline plots in **BEAST** v. 1.8.2 []. Each analysis was run for 10 million steps, using the HKY + Γ model. We employed a strict molecular clock and a coalescence Bayesian skyline tree prior with either the default 10 grouped coalescent intervals or instead 4 groups where 10 or fewer individual sequences were available from a location. Operators were set to auto-optimize, and parameters were logged every 1000 iterations. All other search parameters were as default. We aimed to generate only information on the relative timing of effective population size changes, so no temporal calibrations were employed. Bayesian skyline plots were plotted using Tracer v. 1.6 []. All populations showed evidence of an expansion towards the present day. Thus, we were able to identify the point in relative time where that constant expansion was initiated [] and we refer to this as the time of population establishment. Occasionally, populations showed declines in population size after a period of population establishment and growth. Here, we only use the point of the initial population expansion as the time of population establishment.
Figure 1.We estimated the distance of each of the 314 sampled populations to the centre of the Coral Triangle (). For each species, we fitted a regression line to the relationship between the distance from the centre of the Coral Triangle and the relative time of population establishment. Data for both variables were standardized (mean = 0 and s.d. = 1), ensuring a regression intercept of zero for all species and enabling the calculation of a standardized slope for each species. These slopes were compared to an expected mean slope of zero using a one-sample t-test and bias in the direction of slopes was tested using a binomial test. We also tested if populations undergoing the earliest expansions were relatively closer to the centre of the Coral Triangle than those undergoing the latest establishments using a paired t-test. […]

## Pipeline specifications

Software tools | Clustal W, DAMBE, BEAST |
---|---|

Application | Phylogenetics |