Burns & Broders – Correlates of dispersal extent predict the degree of population genetic structuring in bats
Appendix A. Equations for calculations of FST and details of collection of predictor variables from literature
Calculation of FST
Many estimates of FST were directly reported in studies and used as such in the analysis, or we calculated these values from genetic data presented in tables (e.g. allele frequency tables). When required, we calculated pairwise per locus estimates of FST from allele frequency tables following Nei’s approach [1]: FST = (HT-HS)/HT. HT is the total heterozygosity in the population and was calculated as HT = 1-where is the frequency of allele averaged over the subpopulations and is the number of alleles in the total population. HS is the heterozygosity in the subpopulation s and HS was calculated as HS = where is the frequency of the ith allele in the subpopulation. Per locus FST estimates were then averaged over all loci for a global FST estimate [2].
To facilitate comparison among all types of studies, we converted mitochondrial FST estimates to bi-parentally inherited, diploid gene flow, assuming Wright’s island model estimates sensu Bradbury et al. [3], using the formula of Kinlan and Gaines [4]. This conversion assumes there is an equal sex ratio and equal migration occurring. First, mitochondrial estimates of gene flow were calculated as: ). These values were then multipled by 2 to account for biparental inheritance of nuclear markers, and were recalculated assuming an island model of migration with diploid gene flow as: . Mitochondrial studies which presented ΦST were not included in the analysis because this measure incorporates additional information on substitution differences [5] that cannot be corrected for in the same manner.
When comparing FST across multiple species and studies, the extent of geographic sampling should be accounted for as it can strongly influence the measure of FST[2, 6]. This was done by calculating the maximum distance (DMAX) between sampled sites for each study. Location coordinates were taken directly from the studies or were approximated from maps and subsequently put into Google Earth © vs. 7.0.2.8415 to estimate the coordinates. Straight line maximum distances were then calculated from coordinates and maximum distances were log transformed to meet normality assumptions, and then regressed against estimates of FST using simple linear regression with the residuals used in further analysis.
Details of dispersal extent predictor variables
To calculate the area of the geographic range, shape files of mapped areas were downloaded from the IUCN database [7], polygons with a presence value of 1(Extant) were extracted, and the areas were calculated and summed to represent the range in square kilometers. Geographic range sizes were log transformed to meet normality assumptions for analysis. To characterize the biogeographic range of each species, the maximum and minimum latitudes were calculated from the shape files and the median value (absolute value) was calculated.
For species where no published values of wing loading or aspect ratio could be located (n=5), both measures were estimated from the average body mass of the species using morphological data from the published literature or from the CRC Handbook of Mammalian Body Masses [8]. Specific equations for family or subfamily from Norberg and Raynor ‘s ecomorphology and flight in bats study [9] were used to calculate unknown values. Body mass (M) was entered in kilograms, and calculated wing loading units are N m-2, where N = Newtons (mass times the gravitational constant). Aspect ratio is a unit-less measure derived from the formula wing span area squared divided by wing span area. For a study to be included, measures of all 5 species variables had to be obtained to perform the multivariate analysis.
Table A1. Family or subfamily specific equations for calculating wing loading and aspect ratio for bats from body mass (M) in kilograms. Equations are from Norberg and Raynor (1987).
Family/subfamily / Wing loading / Aspect ratioFamily Pteropodidae , subfamily Pteropodinae / 45.94M 0.339 / 5.40M -0.111
Family Rhinolphidae / 91.41M0.541 / 11.14M 0.128
Family Phyllostomidae, subfamily Glossophaginae / 63.13M0.540 / 13.32M 0.160
Family Vespertilionidae, subfamily Vespertilioninae / 108.1M0.540 / 16.30M 0.191
Family Miniopteridae / 803.0M0.985 / 38.35M 0.362
References
1.Nei M. 1977 F-statistics and analysis of gene diversity in subdivided populations. Annals of Human Genetics, London41, 225.
2.Burland T.M., Worthington Wilmer J. 2001 Seeing in the dark: molecular approaches to the study of bat populations. Biological Reviews76, 389-409.
3.Bradbury I.R., Laurel B., Snelgrove P.V.R., Bentzen P., Campana S.E. 2008 Global patterns in marine dispersal estimates: the influence of geography, taxonomic category and life history. Proceedings of the Royal Society of London Series B275, 1803-1809.
4.Kinlan B.P., Gaines S.D. 2003 Propagule dispersal in marine and terrestrial environments: a community perspective. Ecology84(8), 2007-2010.
5.Excoffier L., Smouse P.E., Quattro J.M. 1992 Analysis of molecular variance inferred from metric distances among DNA haplotypes: Application to human mitochondrial DNA restriction data. Genetics1311, 479-491.
6.Stevens V.M., Turlure C., Baguette M. 2010 A meta-analysis of dispersal in butterflies. Biological Reviews85, 625-642.
7.IUCN. 2012 IUCN Red List of Threatened Species. Version 2012.2 (
8.Silva M., Downing J.A. 1995 CRC handbook of mammalian body masses. Boca Raton, FL, CRC Press.
9.Norberg U.M., Raynor J.M.V. 1987 Ecological morphology and flight in bats (Mammalia;Chiroptera): wing adaptations, flight performance, foraging strategy and echolocation. Philosophical Transactions of the Royal Society of London Series B, Biological Sciences316(1179), 335-427.