Caudata macrogenetics: Geographic attributes and lineage age predict global patterns of mitochondrial genetic variation in salamanders

Caudata macrogenetics: Geographic attributes and lineage age predict global patterns of mitochondrial genetic variation in salamanders | bioRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-M677548'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search New Results Caudata macrogenetics: Geographic attributes and lineage age predict global patterns of mitochondrial genetic variation in salamanders View ORCID Profile Luis Amador , View ORCID Profile Daniele L.F. Wiley , View ORCID Profile Irvin Arroyo-Torres , View ORCID Profile Chris X. McDaniels , View ORCID Profile Esteban O. Rosario-Sanchez , Hannah Farmer , View ORCID Profile Hannah Bradley , View ORCID Profile James Erdmann , Tara A Pelletier , View ORCID Profile Lisa N Barrow doi: https://doi.org/10.1101/2025.01.24.634750 Luis Amador 1 University of New Mexico Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Luis Amador Daniele L.F. Wiley 1 University of New Mexico Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Daniele L.F. Wiley Irvin Arroyo-Torres 1 University of New Mexico Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Irvin Arroyo-Torres Chris X. McDaniels 1 University of New Mexico Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Chris X. McDaniels Esteban O. Rosario-Sanchez 1 University of New Mexico Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Esteban O. Rosario-Sanchez Hannah Farmer 2 Radford University Find this author on Google Scholar Find this author on PubMed Search for this author on this site Hannah Bradley 1 University of New Mexico Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Hannah Bradley James Erdmann 1 University of New Mexico Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for James Erdmann Tara A Pelletier 2 Radford University Find this author on Google Scholar Find this author on PubMed Search for this author on this site Lisa N Barrow 1 University of New Mexico Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Lisa N Barrow For correspondence: lnbarrow{at}unm.edu Abstract Full Text Info/History Metrics Data/Code Preview PDF Abstract Aim Genetic diversity contains valuable information about ecological and evolutionary aspects of species. Intraspecific genetic variation is shaped by species’ natural history traits and by characteristics of geography and climate within their ranges. Amphibians are of ecological and conservation interest because of their global distribution, deep history, trait diversity, and roles within ecological communities. Here, we studied genetic variation within salamanders to investigate predictors of nucleotide diversity and spatial patterns of genetic differentiation. Location Global. Time Period Present. Major Taxa Studied Salamanders. Methods We repurposed mitochondrial DNA sequences and ecological data from open-access databases for 220 salamander species. We calculated nucleotide diversity (π) and tested for isolation by distance (IBD) and isolation by environment (IBE). We analyzed these three variables with random forest and phylogenetic comparative methods using 28 predictors expected to be associated with genetic variation. Results We recovered 8,108 Cytb sequences with their associated geographic coordinates, of which 7,007 sequences were manually curated by us. Range size, lineage age, and sample size were important predictors of genetic variation. We found higher diversity in regions including the Neotropics and central-eastern Europe. The absence of phylogenetic signal in π, IBD, and IBE suggests that genetic variation is shaped by local ecological and geographical factors rather than by shared ancestry. Main Conclusions Our finding of range size as an important predictor aligns with theoretical expectations that species with larger ranges tend to harbor more genetic diversity. Furthermore, lineage age being an important predictor is in line with the clade-age hypothesis, in which species with longer divergence times have higher genetic diversity because they have had more time to accumulate genetic variation. Our results underscore the importance of integrating spatial data into macrogenetic studies, providing valuable information for future studies and conservation strategies targeting regions with high or low genetic diversity. Introduction Genetic variation plays a foundational role in maintaining biological diversity ( Kardos et al., 2021 ). One significant evolutionary factor that predicts genetic variation within populations is the number of individuals that contribute DNA to the next generation (the effective population size; Wright, 1931 ). Genetic variation is proportional to population size, as explained by the neutral theory of molecular evolution ( Kimura, 1968 ). Gene flow between populations is another key factor influencing variation because it can add and maintain novel alleles, counteract the effects of genetic drift and selection, and enable population and species persistence ( Park et al., 2024 ). Natural selection also has an important role in affecting neutral diversity of populations through background or linked selection ( Corbett-Detig et al., 2015 ). Understanding this variation and which factors influence it across different geographic and environmental contexts is essential for elucidating population dynamics and will help us identify populations with conservation needs ( Pauls et al., 2013 ). For example, declining levels of genetic variation may be further exacerbated under global change scenarios due to the reduction of suitable habitat for species ( Schierenbeck, 2017 ). Climatic and geographic factors shape patterns of dispersal and population size through mechanisms such as habitat connectivity, environmental stability, and barriers to gene flow. These processes, in turn, determine the extent and distribution of genetic variation within species (e.g., Hanson et al., 2017 ). For example, geographic distance can limit gene flow among populations, resulting in patterns of isolation-by-distance (IBD), where genetic dissimilarity increases as populations become more spatially separated ( Wright, 1943 ). Similarly, environmental differences can restrict gene flow by selecting against migrants non-adapted to the new environmental conditions, leading to isolation-by-environment (IBE), which reflects the influence of habitat heterogeneity and ecological gradients on genetic variation ( Wang and Bradburd, 2014 ). Historical climatic fluctuations and complex topography often act together to promote cycles of population contraction and expansion, which generate and maintain genetic variation. In the alpine toad Scutiger ningshanensis , for instance, repeated glacial cycles during the Pleistocene, combined with uplifts of mountain ranges, likely fragmented populations and created refugia, increasing opportunities for genetic differentiation ( Meng et al., 2014 ). Life history traits, particularly those related to reproduction, also influence genetic diversity by affecting effective population size and gene flow ( Ellegren and Galtier, 2016 ). For example, species with high fecundity, widespread dispersal, or external fertilization can maintain large and connected populations, which reduces the effects of genetic drift and promotes diversity. Broad-scale analyses across hundreds of plant and animal species further support the idea that reproductive traits and climatic variables jointly shape genetic diversity by influencing demographic stability and dispersal potential ( De Kort et al., 2021 ), suggesting that these variables act as consistent predictors of genetic variation across taxa. Studies on focal taxa within a limited geographic range provide the resolution needed to understand local genetic patterns and idiosyncrasies (e.g., Fonseca et al., 2021 ). It is equally important to investigate genetic variation on broader geographic and taxonomic scales, to have a global perspective of the role of ecological and geographic factors on evolutionary patterns. Macrogenetics is an approach to analyze open-access DNA sequences across large scales and investigate patterns of intraspecific genetic diversity across higher taxonomic groups ( Blanchet et al. 2017 ). Macrogenetic research has gained popularity over the past decade, driven by the increasing availability of molecular resources (e.g., GenBank), computational tools (e.g., machine learning), and trait databases ( Leigh et al., 2021 ). Despite limitations, particularly the genetic markers represented by available data (see Millette et al., 2021 ; Paz-Vinas et al., 2021), efforts to understand large-scale patterns and predictors of intraspecific genetic diversity have been undertaken across various taxonomic groups. A study of global genetic diversity in insects attributed high levels of mitochondrial DNA (mtDNA) genetic diversity in the subtropics to temperature and climatic stability ( French et al., 2023 ). Another study of nucleotide diversity using mtDNA for animals and chloroplast DNA for plants in over 38,000 species found that latitude significantly correlates with genetic diversity, with tropical species exhibiting higher intraspecific diversity ( Fonseca et al., 2023 ). These studies highlight the potential of macrogenetics to uncover important predictors of intraspecific genetic diversity across broad sets of taxa globally. Amphibians contain remarkable diversity in life history strategies and the geographic and climatic variability of their habitats. Recent studies of amphibian genetic diversity indicate that biogeographic region can influence which factors predict intraspecific genetic variation. For example, in Nearctic amphibians, taxonomic family, the number of sequences, and latitude were key predictors of intraspecific variation ( Barrow et al., 2021 ), whereas range size, elevation, latitude, and precipitation predicted genetic variation in Neotropical amphibians ( Amador et al., 2024 ). In the Americas, Lawrence et al. (2023) found a strong relationship between environmental variables and amphibian expected heterozygosity ( H E ), and that amphibian H E decreased with sample size (number of individuals). Within amphibians, salamanders are particularly valuable for macroevolutionary studies because of their broad distribution (excluding Oceania, the Afrotropics, and Antarctica), manageable number of species, and unique traits. They display remarkable ecological and life history diversity that can directly shape population dynamics and, consequently, genetic structure. For instance, body size variation, from the tiny Thorius arboreus (∼20 mm) to the 1.8-meter-long Andrias japonicus, can influence dispersal ability and home range size, thereby affecting gene flow and population structure. Likewise, salamanders include generalist species like the tiger salamander, Ambystoma tigrinum , versus habitat specialist species adapted to aquatic, terrestrial, or arboreal environments, which can promote or constrain connectivity across landscapes. One trait that may influence patterns of genetic diversity in salamanders is developmental mode. Species with direct development (i.e., without a larval stage) often have limited dispersal, which could lead to smaller, isolated populations favoring population differentiation ( Paz et al. 2015 ; Liedtke et al. 2022 ). In contrast, species with a larval stage and dispersing terrestrial adults may maintain more interconnected populations, reducing genetic differentiation (e.g., Zamudio and Wieczorek 2007 ). Some salamanders exhibit paedomorphosis, a reproductive strategy in which adults retain larval characteristics, which reduces gene flow by allowing persistent aquatic life cycles in species that could otherwise be terrestrial ( Denoël et al., 2005 ). Recent studies have begun to untangle the relationships between species-level traits and genetic variation. For example, Segovia-Ramírez et al. (2023) used genomic data from 62 Neotropical salamander species (one individual per species) to demonstrate that precipitation variability and snout-vent length (body size) were significant predictors of genomic diversity, perhaps related to their roles in shaping population persistence and abundance. Parsons et al. (2024) combined genetic, geographic, climatic, and life history data from salamanders and used a machine learning model to identify predictors of unrecognized genetic lineages. They found that Caudata hidden diversity is the result of variation in climatic variables. A more comprehensive, georeferenced view of intraspecific genetic variation in salamanders, encompassing more species and potential predictors (e.g., climatic and natural history traits), would provide useful insights into the evolutionary and demographic processes driving variation in this group. Here, we combine molecular, phenotypic, and environmental open-access data to investigate the determinants of intraspecific genetic variation in salamanders. We used machine learning techniques and phylogenetic comparative methods to determine whether natural history traits or climatic or environmental variables can predict genetic variation within species on a global scale. We hypothesized that reproductive traits, such as the total number of eggs (a proxy for fitness) and reproductive mode (e.g., larval vs. direct development), would predict intraspecific genetic variation in salamanders. Species that produce more offspring may maintain larger, more stable populations, which can preserve higher levels of genetic diversity over time ( Lacy, 1987 ; Reed and Frankham, 2003 ). Additionally, reproductive mode could influence dispersal, leading to lower expected connectivity and more genetic differentiation in species with direct development compared to those with aquatic larvae. We also predicted that traits associated with total species population size (e.g., range size) would be positively correlated with genetic variation, as larger ranges may support more numerous and genetically diverse populations ( Frankham, 1996 ). We hypothesized that species with older evolutionary age (lineage age) may have higher genetic diversity because they have had more time to accumulate variation than younger species (the clade-age hypothesis; see McPeek and Brown, 2007 ; Scholl and Wiens, 2016 ). Finally, we expected high spatial autocorrelation in nucleotide diversity of salamanders overall based on their limited dispersal capabilities, which can lead to genetic clustering at small scales ( Fusco et al., 2021 ). Methods Data collection We followed the taxonomic classification proposed by AmphibiaWeb ( amphibiaweb.org ) and Amphibian Species of the World ( https://amphibiansoftheworld.amnh.org/ ) and prepared a detailed count of all species included in this study (Supplementary Material 1, SM1). These databases were also used to reconcile species names and verify taxonomic consistency across datasets. When discrepancies or outdated names were found, we updated species names to match current taxonomy, with a description of our decisions in SM1. We repurposed Caudata DNA sequences, natural history traits, and geographic and environmental data from open-access databases. Following the approach of recent research on the determinants of genetic diversity of multiple amphibian species (e.g., Barrow et al., 2021 ; Amador et al., 2024 ), we assembled sequences of the cytochrome-b (Cytb) mitochondrial gene. This gene is suitable for genetic diversity studies because it contains both conserved and variable regions, is commonly used in amphibian single-locus phylogenetic and population genetic studies, and is therefore the most abundant gene in open-access databases for amphibians. Sequences were obtained from GenBank (National Center for Biotechnology Information, NCBI), phylogatR ( Pelletier et al. 2022 ), the phruta R package ( Román Palacios, 2023 ), and Amador et al. (2024). We included species with at least 10 sequences, seeking a better representation of intraspecific genetic variation. Our dataset contained species with high variation in sampling effort, such as those with more than 200 sequences. To account for this sampling bias, we also conducted a set of analyses on randomly subsampled datasets that included 10–20 sequences per species. Alignments were generated, visualized, and edited with the MUSCLE aligner v.3.8.31 ( Edgar, 2004 ) in AliView v.1.28 ( Larsson, 2014 ) and saved in FASTA format. The geographic coordinates (latitude and longitude in decimal degrees format) previously associated with each Cytb sequence were recovered from GenBank, phylogatR, and Amador et al. (2024); corresponding to 31 species (14.1% of the total species) and 1,169 sequences (14.3% of the total sequences). We then manually retrieved the remaining geographic coordinates associated with the sequences (corresponding to 189 species and 7,007 sequences) following a tutorial prepared for this purpose (see Supplementary Material 2). We georeferenced localities using either the function geocode() of the tidygeocoder R package ( Cambon et al., 2021 ) or the GEOLocate web application ( Rios and Bart, 2010 ). We only retained sequences that were georeferenced with an error of less than 20 km, resulting in 12 species with fewer than 10 georeferenced sequences. We calculated both the total range size and the sampled range size of DNA sequences for each species as follows. For total range size, we obtained the geographic range map in shapefile format for each species from the International Union for Conservation of Nature (IUCN) portal using the getIUCN() function of the rasterSp R package ( RS-eco, 2023 ). Species ranges not available in the IUCN portal (24 species) were generated with minimum convex polygons using the sequence coordinates and the function mcp() in the R package adehabitatHR ( Calenge, 2024 ; see Supplementary Material). This same approach was also used to obtain the geographic range of the sampled sequences for all species (i.e., using only the sequence occurrences). Range size was calculated from shapefiles using the areaPolygon() function of the geosphere R package ( Hijmans, 2022 ). Elevation and latitude (min, max, and mean) were obtained from the retrieved geographic coordinates using the function get_elev_point() of the elevatr R package ( Hollister et al., 2025 ) for elevation, and custom R scripts for latitude (see Supplementary Material 3). We also obtained precipitation and temperature data (mean and standard deviation) from WorldClim v2.1 ( Fick and Hijmans, 2017 ) using the function worldclim_global() of the geodata R package v0.6-2 ( Hijmans et al., 2024 ). We used the same natural history and geographic traits as in Amador et al., (2024) with slight additions and omissions, choosing predictors we expected to be related to genetic variation (e.g., body size, reproductive mode, elevation, latitude, species range size, precipitation and temperature; Table S1). We omitted the variable activity because nearly all salamanders (except three of the 220 species) included in our dataset were nocturnal. Eight new predictors were added to the matrix (28 predictors in total), which included sampling effort/coordinates (i.e., number of sequences with associated coordinates), sequence length/coordinates (i.e., number of base pairs of sequences with associated coordinates), snout-vent length in millimeters (SVL), paedomorphism (whether or not juvenile features are retained as an adult), litter size min (minimum number of offspring or eggs per clutch), litter size max (maximum number of offspring or eggs per clutch), latitudinal midpoint (mean of the minimum and maximum latitudinal values), and lineage age (estimated species divergence times obtained from Stewart and Wiens, 2025 ) (Table S1). We obtained trait information from the open-access database AmphiBIO ( Oliveira et al. 2017 ), species accounts in AmphibiaWeb ( AmphibiaWeb, 2024 ), and relevant literature. Measures of genetic variation We evaluated overall genetic diversity (range-wide nucleotide diversity) and spatial patterns of genetic variation (IBD and IBE) within salamander species. We calculated nucleotide diversity (π) with the function nuc.div() of the pegas R package ( Paradis, 2010 ) for all sequences (π A ) and sequences with associated coordinates only (π S ). Genetic distances ( gendist ) for each species were calculated based on the raw difference between sequences, using the dist.dna() function of the ape R package ( Paradis and Schliep, 2019 ). We tested IBD and IBE for each salamander species as follows. Geographic distance ( geodist ) was calculated between each pair of sequence coordinates using Euclidean distance with the dist() function of the stats package ( R Core Team, 2023 ) in R. Before calculating environmental distances, we used the function scale() in R to standardize variables measured in different units (e.g., temperature – °C or precipitation-mm). Environmental distance ( envdist ) was calculated between each pair of sequence coordinates based on the 19 variables of the WorldClim v2.1 database using the vegdist() function of the vegan R package ( Oksanen et al., 2024 ). This function calculates a distance matrix, selecting all environmental variables using Euclidean distances. We performed Multiple Matrix Regression with Randomization (MMRR) analyses for each species ( Wang, 2013 ). MMRR quantifies the relative effects of IBD and IBE, represented as distance matrices ( geodist and envdist , respectively), in explaining genetic variation, represented by genetic distances. We compared the observed correlations between geodist vs. gendist (IBD) and envdist vs. gendist (IBE) to the correlations from 1,000 permutations. Species were coded as ‘Yes’ if they did present IBD and IBE or ‘No’ if they did not based on an assessment of the associated p-values from the permutation tests. To account for multiple comparisons, we applied a strict Bonferroni correction ( Bonferroni, 1936 ) to reduce the p-value for assessing significance. We used the output (yes or no IBD; yes or no IBE) as binary response variables for further analyses. We mapped π S to visualize the distribution of genetic diversity across the globe and assess differences among biogeographic realms (Nearctic, Neotropic, Oriental, and Palearctic). Briefly, following Amador et al. (2024) , we combined sequences within a species based on a maximum distance of 100 km between coordinates, and calculated π S for each locality (see Supplementary Material 4). We then mapped π S for four different grid sizes (1–4), which correspond with one to four decimal degree grids, or approximately 110 km 2 to 440 km 2 . We also visualized the number of sequences per grid cell for the four resolutions. For these visualizations, we used the R packages ggplot2 ( Wickham, 2016 ), ggspatial ( Dunnington, 2023 ), and sf ( Pebesma, 2018 ; Pebesma and Bivand, 2023 ). Finally, we tested for spatial autocorrelation for π and the number of sequences separately with the Moran’s I test using the moran.test() function in the spdep R package ( Bivand et al., 2013 ). To further understand the potential link between historical demography and the patterns of genetic variation we found, we used a common statistic for evaluating population size change, Tajima’s D ( Tajima, 1989 ), using the tajima.test() function of the pegas package. This test is compatible with different interpretations related to natural selection versus demographic change, making it difficult to distinguish among hypotheses ( Yang, 2014 ). Therefore, we evaluate these results for comparison rather than include the statistic as a variable in our primary analyses. We followed Fonseca et al. (2023) in assuming that the demographic change interpretation is likely more relevant for global comparisons. Thus, in our study, we interpret Tajima’s D values close to zero as representative of neutrality (no population size change over time), positive values as evidence of recent population contraction, and negative values as recent population expansion. Predicting genetic variation using random forests (RF) We used random forest (RF) regression to predict π and RF classification to predict IBD and IBE based on a set of geographic range characteristics and ecological variables. We built RF models with the function randomForest() in the randomForest R package ( Liaw & Wiener, 2002 ), with 10,000 trees and 100 permutations per model. We used the tuneRF() function to obtain the optimal mtry parameter (number of variables that are randomly sampled as candidates at each split) and then ran a new RF regression or classification analysis using the best mtry value. The data was split into a training set (70% of the data) for building models and a testing set (30%) for making predictions. For RF regression, we assessed models using the R-squared (R 2 ) statistic, which represents the proportion of the variance explained by the model, with a higher R 2 value indicating a better fit and more predictive power. The relative importance of the predictor variables for π A and π S was evaluated based on the percentage increase in mean squared error (%IncMSE). In RF classification, in addition to the original models, we also tried upsampling and downsampling to address the large class imbalance within our datasets (yes IBD = 56 species, no IBD = 161 species; yes IBE = 39 species, no IBE = 178 species; based on MMRR results with Bonferroni correction). For downsampling, the majority cases (no IBD/IBE) were randomly subsampled to equal the number of minority cases (yes IBD/IBE). For upsampling, minority cases were randomly duplicated to equal the number of majority cases. We chose the best model based on the lowest out-of-bag (OOB) error rate. To assess overfitting and improve cross-validation in the model, we split our data into a training set and a test set (70% train, 30% test). To evaluate predictions of the RF classification models, we created a confusion matrix with the confusionMatrix() function of the caret R package ( Kuhn, 2008 ) and tested whether model accuracy was significantly better than the no information rate (NIR). Finally, we assessed the relative importance of the variables based on the Mean Decrease Accuracy (MDA) for the RF classification models of IBD and IBE. Evaluating predictors of genetic variation with phylogenetic comparative methods For phylogenetic analyses, we downloaded a phylogeny subset from VertLife.org for the species in our dataset, which represents phylogenetic relationships inferred by Jetz and Pyron (2018) . We used a subset of 180 species for these analyses because 40 species from our dataset are not included in the available phylogeny. Using a custom R script (Supplementary Material 3), we edited the name of 12 species in the tree to match the names in the dataset (e.g., some Speleomantes and Hypselotriton species). To assess the potential effect of phylogenetic history in our models, we tested for phylogenetic signal in π using Pagel’s λ (lambda; Pagel, 1999 ) and Blomberg’s K ( Blomberg et al., 2003 ) implemented in the phytools R package ( Revell, 2012 ) with the function phylosig(). For IBD and IBE, we used the D statistic implemented in the package caper with the function phylo.d() to assess phylogenetic signal. We used phylogenetic independent contrasts (PIC; Felsenstein, 1985 ) to investigate the evolutionary correlation between π and the most important predictors identified in the RF analyses. We used the pic() function of the ape R package to compute PIC for both response and explanatory variables. We then fit linear models in R for both the uncorrected and phylogenetically corrected (PIC) response and predictor variables. We also used phylogenetic generalized linear mixed models (PGLMMs) to analyze binary traits as described in Ives and Helmus (2011) . PGLMMs allow us to account for phylogenetic covariance while reducing Type I error rates. This method was used to analyze IBD and IBE as binary dependent variables (0 for no IBD/IBE, 1 for yes IBD/IBE) versus the most important predictors based on RF results. Because of the nature of our data (binary outcomes), we chose binomial as the error family with a logit link function. We compared IBD vs. sampling range size, number of sequences, maximum latitude, and mean precipitation; and IBE vs. sampling range size, number of sequences, mean elevation, and mean latitude. We implemented this analysis using the function pglmm() in the phyr R package ( Li et al., 2020 ), including a phylogenetic random effect to account for the non-independence among species due to shared evolutionary history. Results Data summary We obtained 12,961 Cytb sequences from 220 species distributed globally, of which 8,937 sequences were retrieved directly from GenBank, 2,349 sequences from phylogatR, 1,657 from the phruta package, and 296 from Amador et al. (2024) . Of these, we obtained geographic coordinates for 8,108 sequences from 219 species. A single species, Plethodon sherando , did not have associated coordinates, therefore was not included in IBD and IBE analyses; and 11 species had between four and nine sequences with coordinates. The final dataset included 220 species in nine families (representing >26% of species and 90% of families globally), three response variables (π, IBD, and IBE), and 28 predictors (Table S2). All sequences associated with geographic coordinates and R scripts used to retrieve sequences and coordinates are available as Supplementary Material 4. Patterns of genetic variation in global salamanders Nucleotide diversity within species ranged from π A = 0 to π A = 0.087, with Pseudoeurycea lineola exhibiting the highest diversity when all sequences were analyzed. When considering only sequences with associated coordinates, values ranged from π S = 0 to π S = 0.089, with Desmognathus amphileucus showing the highest diversity. We found that nucleotide diversity for all sequences (π A ) and for the sequences with associated coordinates (π S ) were highly correlated (Figure S1; Pearson’s correlation r = 0.95, t (218) = 46.18, p-value < 0.001). Based on this validation, we conducted further analyses using sequences with associated coordinates (π S ). Bolitoglossa yariguiensis was the only species with a value of π = 0 (Table S2; Figure S2). We found that π and the number of sequences were not randomly distributed across the world map ( Figure 1 , Figures S3 – S5). For example, regions with high genetic diversity included the Ecuadorian Amazon and central-southern Mexico in the Neotropics; and central-eastern Europe in the Palearctic. Areas with low genetic diversity were identified in northern North America (Nearctic), western Europe, and central-western China in the Oriental realm ( Figure 1 ). Download figure Open in new tab Figure 1. Maps of A) mean nucleotide diversity (π) and B) mean number of sequences with coordinates (log scale) of global salamanders. Biogeographic realms are indicated with shading: Nearctic (brown), Neotropic (green), Oriental (blue), and Palearctic (purple). Boxplots of nucleotide diversity for each biogeographic realm are shown in map A. The number of species and mean number of sequences with coordinates for each biogeographic realm are shown in map B. For visualization purposes, we show grid cell = 4° (approximately 440 km 2 ) for mapping. Furthermore, we found statistically significant, but weak, positive spatial autocorrelation in both π S (Moran’s I = 0.1435, p-value < 0.0001) and the number of sequences (Moran’s I = 0.4136, p-value < 0.0001). Most species did not present IBD and IBE patterns, with 161 species (74.2%) showing no IBD and 178 species (82%) showing no IBE (Table S2; Figure S6). We found that only 21 species had significant patterns for both IBD and IBE (Table S2). The Neotropical and Oriental realms had the lowest percentages of species that showed IBD and IBE (IBD: 13.3% and 24.4%, IBE: 13.3% and 10.8% respectively; Figure S7). The Palearctic region had the highest percentage of species that showed IBD (31%) and IBE (25.4%) patterns (Figure S7). Genetic variation was randomly distributed across the phylogeny ( Figure 2 ). For example, we found species in the same genus with dissimilar π values, such as Plethodon (range = 0.0008 – 0.0698) and Eurycea (range = 0.0015 – 0.0768) in Plethodontidae, Paramesotriton (range = 0.001 – 0.0767) in Salamandridae, and Hynobius (range = 0.0008 – 0.0853) in Hynobiidae. We found no or low phylogenetic signal in π based on the method used, for example, with Pagel’s λ = 0.0005, p-value (based on LR test) > 0.05; and Blomberg’s K = 0.053, p-value (based on 1000 randomizations) > 0.05 (Figure S8). We found no significant phylogenetic signal in IBD (Estimated D IBD = 0.945, p-value = 0.249), and significant but moderate phylogenetic signal in IBE (Estimated D IBE = 0.782, p-value = 0.012) (Figure S9). Download figure Open in new tab Figure 2. Phylogeny of Caudata species included in this study, showing the values of nucleotide diversity (π) as gray barplots to the right and the presence (brown) or absence (teal) of isolation-by-distance (IBD) and isolation-by-environment (IBE) per species. Relationships are based on Jetz et al. (2018). Notophthalmus viridescens silhouette was taken from phylopic.org. Our results from assessing Tajima’s D indicate that many species in our dataset (n = 109) have undergone recent population expansion (i.e., negative Tajima’s D, more recent mutations than expected under neutrality; Figure S10). Positive Tajima’s D values indicate more alleles at intermediate frequencies than expected under neutrality, possibly because of a recent population contraction or bottleneck, as we observed in 51 salamander species (Figure S10). Fifty-nine species do not show strong evidence for population expansion or contraction (Tajima’s D ≈ 0) suggesting no deviation from expectations under neutrality (Table S2; Figure S10). Predictors of genetic variation in global salamanders Our random forest analyses showed an influence of sampling effort on genetic variation estimates, with the number of sequences (with associated geographic coordinates) as a top predictor of IBD, and IBE ( Figure 3 ). Additionally, we compared both species range size variables in our dataset, total range size versus sampling range size, and we found a strong, positive linear relationship (Figure S11; Pearson’s correlation r = 0.87, t (218) = 25.82, p-value < 0.001). These results suggest the sampled range for most species adequately represented the total range size. Nevertheless, we ran RF analyses using both total range size or sampling range size and found similar results between the two variables ( Figures 3 , S12). Because the range size variables were highly correlated with one another and captured similar aspects of the species’ biology, we subsequently describe the results using the sampling range size. Download figure Open in new tab Figure 3. Random Forest (RF) results for A) nucleotide diversity (π, using RF regression), B) isolation-by-distance (IBD, using RF classification), and C) isolation-by-environment (IBE, using RF classification). All predictors except the number of base pairs, and the total range size were used in these RF models, but only the top 10 best predictors for each model are shown here. Range size* = sampling range size based on localities with associated DNA sequences; # of sequences + = number of sequences associated with geographic coordinates; SVL = snout-vent length; mm = millimeters; BIO1 = Annual Mean Temperature; BIO4 = Temperature Seasonality (standard deviation ×100); BIO12 = Annual Precipitation; BIO15 = Precipitation Seasonality (Coefficient of variation). We found that sampling range size and lineage age were the best predictors of π and were two of the best predictors of IBD and IBE in random forest analyses ( Figure 3 ). Random forest regressions showed that taxonomic family was the fourth-best predictor of π following mean temperature (BIO1) considering the Increase in Mean Squared Error (%IncMSE) measure ( Figure 3a , S13). Validation of the RF regression results for π based on the R 2 score indicated good model performance (R 2 = 0.95). Mean precipitation (BIO12) and lineage age were other important predictors of IBD ( Figure 3b , S14). For IBE, RF showed that latitude was another important predictor ( Figure 3c , S15). In RF classification analyses, the best IBD and IBE models were obtained when we downsampled our dataset. Our IBD RF model had high accuracy (88.1%; 95% Confidence Interval CI: 77.82% – 94.7%), low sensitivity (56.25%), and very high specificity (98.04%); our model was statistically better than predicting only the majority class, p-value [Acc > NIR] = 0.0116. The IBE RF model had high accuracy (94.74%; CI: 85.38% – 98.9%), low sensitivity (57.14%), and very high specificity (100%); however, the model’s accuracy was not statistically better than the NIR, p-value [Acc > NIR] = 0.0688. Sampling range size had a significant, positive relationship with π, showing that genetic diversity within species increases with the geographic area sampled ( Figure 4A ; R 2 = 0.09; p-value < 0.001). We found a similar pattern using phylogenetic independent contrasts (PIC), where sampling range size and π had a strong, positive relationship ( Figure 4B ; R 2 = 0.11; p-value < 0.001). The relationship between lineage age and π was positive and significant, in which older species had higher genetic diversity in comparison with younger ones (Figure S16A; R 2 = 0.11, p-value < 0.001), but this correlation was not significant taking the phylogeny into account (Figure S16B; R 2 = 0.004, p-value: 0.405). We found a significant, positive relationship between mean temperature (BIO1) and π, suggesting that species living in areas with higher temperatures have more genetic diversity (Figure S17A). However, PIC did not show a significant correlation between temperature (BIO1) and π after correcting for phylogenetic relatedness (Figures S17B). Download figure Open in new tab Figure 4. A) Comparison of sampling range size (in log scale) with nucleotide diversity (π) for each species. B) Comparison of phylogenetic independent contrast (PIC) results between the sampling range size (log scale) and nucleotide diversity (π). Linear regression results are indicated in each panel. The number of sequences associated with geographic coordinates, lineage age, and sampling range size were top predictors for IBD and IBE models. There were clear differences between the number of sequences sampled and IBD/IBE patterns (p-value < 0.01), in which species with more sequences tended to show significant IBD/IBE patterns ( Figure 5A , D). Species with significant IBD/IBE had slightly larger sampling range sizes on average ( Figure 5C , F) and also tended to have older lineage ages ( Figure 5E , S18). When temperate and tropical species were considered separately, we found a similar trend in which species with larger ranges were more likely to exhibit IBD and IBE, but results were not significant (Figure S19). Phylogenetic generalized linear mixed models (PGLMM) found that the number of sequences had a statistically significant, positive relationship with IBD (β = 0.01655, SE = 4.10 x 10 3 , p < 0.001), and IBE (β = 0.0158626, SE = 4.27 × 10 3 , p < 0.001). These results suggest that as sampling effort increases, the likelihood of detecting an IBD or IBE pattern increases (Table S3). Coefficients, standard errors, Z-scores, and p-values of the other predictors (i.e., sampling range size and precipitation for IBD, and lineage age for IBE) are shown in Table S3. Download figure Open in new tab Figure 5. Boxplots comparing the number of sequences (A, D) and the sampling range size (C, F) between species with (Yes, brown color) and without (No, teal color) isolation-by-distance (IBD) and isolation by environment (IBE). Comparisons of precipitation (BIO12) with IBD (B), and lineage age with IBE (E) are also shown. Each panel includes the results of a chi-squared test. Black dots represent outliers. Discussion In this study, we analyzed 12,961 Cytb sequences from 220 global salamander species and identified key predictors of intraspecific diversity (π) and patterns of spatial genetic variation (IBD and IBE). Species range size and lineage age were important predictors of all three aspects of genetic variation. Specifically, species occupying larger geographic areas tended to have greater overall nucleotide diversity and to be more geographically and environmentally structured. The results for π are consistent with population genetics theory ( Wright, 1931 ), in that large geographic ranges can support larger populations that harbor more genetic diversity because of reduced genetic drift ( Lacy, 1987 ). The significant relationship found between range size and spatial genetic differentiation (i.e., IBD and IBE) could be explained by the greater potential for geographic and environmental isolation between populations at distant locations in species with large ranges ( Eckert et al., 2008 ). However, species range size alone does not explain much of the variation in IBD or IBE presence (small effect sizes recovered in logistic regression models), and the strong influence of the number of sequences sampled highlights one of the major challenges for macrogenetic studies—limited data availability for many species worldwide. Our results provide only partial support for the clade-age hypothesis (e.g., random forest results). Although lineage age showed a positive and significant relationship with intraspecific genetic variation across global salamanders, there is no association when the phylogeny is included (see PIC and PGLMM results). For example, we tested whether older salamander lineages tend to accumulate more intraspecific genetic diversity than younger ones, the clade-age hypothesis (see McPeek and Brown, 2007 ), and we found that there is a significant non-phylogenetic correlation, but this relationship disappears once phylogenetic relatedness is taken into account. Similar patterns have been observed in other vertebrate groups, where evolutionary time is recovered as one of the best predictors of genetic diversity (e.g., Theodoridis et al., 2020 in mammals). Our results suggests that the observed effect of lineage age on genetic diversity could be explained by shared evolutionary history rather than independent accumulation of variation through time. The strong phylogenetic signal detected for lineage age (Pagel’s λ = 0.9999, p < 0.001) indicates that this trait is almost entirely structured by shared ancestry. This high phylogenetic dependence reinforces that comparisons involving lineage age must account for shared ancestry (PIC or PGLMM models) to avoid inflated associations with other traits, such as π, IBD or IBE (e.g., Lai et al., 2025 ). Thus, our results indicate that spatial (range size) and environmental factors (precipitation, temperature) may play a stronger role than lineage age in shaping intraspecific genetic variation across Caudata. This suggests that long-term ecological and demographic stability, rather than lineage age, may better explain the maintenance of high intraspecific genetic variation in some salamander lineages (e.g., Pan et al., 2019 ; Iannella et al., 2025 ). The recovered relationship between lineage age and intraspecific genetic variation may be confounded by lineage-level traits (e.g., life history traits, species range size) that are conserved phylogenetically. In fact, we found a positive and significant relationship between lineage age and sampling range size, as recently proposed by Alzate et al. (2025) . Global patterns of nucleotide diversity (π) Our results provide evidence of the uneven distribution of nucleotide diversity (π S ) across the globe. We observed that certain hotspots of genetic diversity, like the Ecuadorian Amazon and central Europe, southwestern and central North America, and southeast Asia, align with regions of high biodiversity (e.g., Mittermeier et al., 2011 ). Contrastingly, low genetic diversity in regions such as northern North America or western Europe could reflect recent recolonizations or historical bottlenecks (e.g., Riberon et al., 2002 ; Bonato et al., 2018 ; Auteri et al., 2022 ). Our results only partially contrasted with those of Miraldo et al. (2016) , who found that amphibian genetic diversity decreased at higher latitudes. Although our map revealed a broadly similar geographic pattern, latitude itself (mean, maximum, minimum, or midpoint) was not an important predictor of genetic diversity in salamanders. Instead, the predictors that best explained genetic diversity, such as species range size and environmental factors, tended to be associated with regions closer to the equator. The high diversity we found in the Neotropics (16 species) could be explained by a combination of factors including tropical climate stability, high topographic heterogeneity, or long evolutionary history in the Neotropics ( Antonelli, 2022 ). Another plausible explanation for the observed high levels of π in some regions is that they coincide with glacial refugia, areas where species persisted and maintained high levels of genetic diversity through time (e.g., Ramírez Barahona and Eguiarte, 2013 ). The Andean foothills were a major glacial refugium, particularly the central Andes in the Neotropical region (e.g., Escobar et al., 2021 ). In the Nearctic, the Southern Appalachians and coastal areas of the Pacific Northwest are considered potential glacial refugia (e.g., Shafer et al., 2010 ). In the Palearctic, the Caucasus Mountain range has been identified as a potential refugium during glacial periods (e.g., Triturus newts; Wielstra et al., 2013 ). The Korean Peninsula and the mountainous regions of China are two examples of glacial refugia in the Oriental region ( Fu and Wen, 2023 ). Low levels of π in northern North America are consistent with the impact of historical and demographic processes in the region, such as Pleistocene glaciations and associated range contractions in salamanders ( Rovito and Schoville, 2017 ; López-Delgado and Meirmans, 2022). We showed that species experiencing expansion, based on Tajima’s D, presented low levels of nucleotide diversity (Figure S20). This result is expected when recent expansion from a small population occurred after a bottleneck, leading to a limited gene pool carried by a few founding individuals, such as range expansions after the last glacial maximum ( Mayr, 2001 ). Interestingly, the occupation of refugia by salamander species may have contributed to the extinction of older lineages and fostered more recent diversification events, consistent with patterns observed in the regions mentioned above (see Tingley and Dubey, 2012 ). We observed that species experiencing demographic expansion often have young evolutionary age (Figure S21), reflecting recent diversification events associated with demographic growth. Global patterns of isolation-by-distance (IBD) and environment (IBE) The limited evidence for both IBD and IBE across species (only 21 species, or ∼10%, showed both patterns) suggests that gene flow in many salamander species may not be constrained by geography or environment. Instead, the genetic variation in these species is shaped more by demographic processes such as bottlenecks or population expansions than current spatial or ecological gradients (e.g., Sexton et al., 2024 ). For instance, some salamander species may maintain high gene flow that can homogenize genetic variation, reducing patterns of IBD or IBE. In contrast, others with strong philopatry may exhibit spatial genetic structure influenced more by historical processes that disrupt patterns of genetic isolation associated with IBD and IBE ( Wake, 2009 ; Sexton et al., 2014 ). For example, montane populations isolated in separate refugia in response to climatic changes might differentiate and exhibit highly restricted gene flow even at short distances or among similar environments, as reported in plethodontid salamanders ( Kozak and Wiens, 2006 ; Rovito, 2017 ). These results underscore the importance of considering biogeographic and ecological context when evaluating the drivers of intraspecific genetic variation on a global scale. The proportions of species showing IBD and IBE differed slightly across biogeographic realms, with the Palearctic including the most species with these patterns. One example is the fire salamander ( Salamandra salamandra ), one of the most common and widespread salamander species in Europe, which was found to be strongly structured in a landscape genetics analysis ( Bani et al., 2015 ). The Neotropical and Oriental regions had lower proportions of species with IBD or IBE patterns compared to the other regions, but these differences were relatively minor. This contrast between realms suggests that historical biogeography and environmental complexity are key to understanding spatial genetic variation. For example, the incidence of IBD/IBE in the Nearctic and Palearctic could reflect the consequence of glacial cycles ( Schmitt, 2007 ), while the low incidence in tropical regions may reflect more recent colonization dynamics that occurred in the late Pleistocene ( Carnaval et al., 2009 ). Predictors of genetic variation in global salamanders Two of the best predictors of nucleotide diversity in this study, the number of sequences and range size, could both be considered proxies of demographic processes. Species with more available sequences could truly be more abundant in nature, reflecting both a true demographic property as well as sampling bias towards species that are more frequently encountered. Either way, sampling more sequences could provide a greater likelihood of detecting genetic variation because more samples analyzed will capture the full range of genetic diversity present within a species. When we analyzed π, IBD, and IBE using a standardized sampling effort per species (10 to 20 sequences), we note that the number of sequences was not an important predictor in π and IBE based on the random forest models, but was the most important predictor of IBD. Similarly, species sampling range size was not a key predictor for IBD and IBE but was an important predictor for π in this subsampled dataset (Figure S22). Lineage age was one of the best predictors for π and IBE in the subsampled dataset (Figure S22). These results suggest that differences in sampling effort across species can provide challenges for macrogenetic studies and should be considered carefully in future studies. The positive correlations found between number of sequences, range size, and π in our dataset are consistent with the prediction that total population size (with number of sequences and range size as potential proxies) corresponds with genetic diversity (Figure S23). Species with larger geographic ranges can have larger and more balanced populations, reflecting the relationship between effective population size and genetic variation ( Waples, 2022 ). Geographic range size was also one of the best predictors of IBD and IBE, like previous findings in a wide range of taxonomic groups ( Pelletier and Carstens, 2018 ). A large range can harbor more differentiation, where population subdivision could be influenced by different ecological and evolutionary pressures and undergo contrasting demographic histories ( Lowe et al., 2017 ). The relationships between range size and patterns of genetic variation we identified in global salamanders is partially consistent with previous regional studies, such as Amador et al. (2024) , which found a similar pattern in Neotropical amphibians for π, but not for IBD or IBE. Similar to previous meta-analyses of IBD and IBE ( Crispo and Hendry, 2005 ; Jenkins et al. 2010 ), we found that identifying global predictors of spatial genetic variation is challenging because of the many potential factors that influence the presence and detection of these patterns. Climatic variables, including different aspects of temperature and precipitation, were among the top ten predictors of genetic variation in RF analyses, suggesting an important role of climate in diversity and population differentiation within salamanders. Mean temperature (BIO1) was the third best predictor for π. The association between higher genetic variation and warmer temperatures is consistent with the evolutionary speed hypothesis ( Wright et al., 2010 ; Gillman and Wright, 2014 ), where genetic diversity is expected to increase with temperature. Variation in precipitation (BIO12) was the second most important predictor for IBD after the number of sequences. A possible explanation for this association is that current stable climates, such as those in tropical environments, promote spatial genetic variation, thus supporting the climate stability hypothesis ( Janzen, 1967 ; Stevens, 1989 ). Although the relationships between these climatic variables and the response variables we examined were relatively weak, previous work has also highlighted the role of climate in explaining variation within salamanders. Specifically, phylogeographic structure measured using species delimitation methods was best explained by variation in climate ( Parsons et al., 2024 ). Like other studies on birds ( Smith et al., 2017 ) and amphibians ( Barrow et al., 2021 ), our analyses suggest that life history and ecological traits are less important than geographic or climatic variation for predicting intraspecific diversity in salamanders. Phylogenetic, taxonomic, and conservation implications We did not find phylogenetic signal in π, although taxonomic family was the fourth-best predictor of π in the final Percentage Increase in Mean Squared Error (%IncMSE) RF model, suggesting that some variation can be explained by evolutionary history, but not all. Sample sizes were limited within some families, but diversity estimates in those families were consistently high (e.g., Sirenidae) or low (e.g., Cryptobranchidae), indicating family can be a useful predictor of intraspecific variation. On the other hand, discrepancies in π within genera of family Plethodontidae, such as observed in Plethodon or Eurycea , emphasize the influence of local ecological and recent demographic processes over shared ancestry for influencing variation within species. Processes leading to homoplasy (e.g., convergence, parallel evolution) dilute phylogenetic signal ( Klingenberg and Gidaszewski, 2010 ). Homoplasy is a common phenomenon in salamander evolution; for example, paedomorphosis has evolved independently several times ( Wake, 2009 ). Likewise, levels of genetic variation may be similar in salamander species that do not share a recent common ancestor ( Kozak and Wiens, 2006 ). Phylogenetic signal in IBD was not statistically significant either, suggesting other factors independent of phylogenetic history influence these patterns. This result was consistent with a previous study on Neotropical amphibians ( Amador et al. 2024 ), which found no phylogenetic signal in IBD or IBE. Significant phylogenetic signal in nucleotide diversity, however, was previously found in amphibian datasets including frogs ( Barrow et al. 2021 ; Amador et al. 2024 ). Species with higher π values could contain cryptic diversity and would be useful target species for exploring species boundaries in future studies (e.g., Parsons et al. 2022 ). Using a smaller dataset than the present study (83 salamander species total, most from the COI mitochondrial gene), Parsons et al. (2024) found that ∼2/3 of species in the dataset showed strong phylogeographic structure, determined by single locus molecular species delimitation methods. Several species with high values of π in our dataset match species from their study, indicating they have hidden genetic lineages that should be further explored ( Mertensiella caucasica, Bolitoglossa rufescens, Plethodon cinereus, Eurycea cirrigera, Batrachoseps attenuatus, Ommatotriton vittatus, Aneides flavipunctatus ), ideally incorporating nuclear genomic data and thorough geographic sampling. Salamanders are within one of the most threatened vertebrate groups ( Luedtke et al., 2023 ) and require innovative syntheses of available information to highlight priorities for conservation strategies. Nucleotide diversity could be a useful proxy for conservation status (Petit-Marty, 2021). The relationship we identified between species range size and genetic variation highlights the vulnerability of species with small ranges. Conservation status was an important predictor for π, but sample sizes were limited within several categories. Seventy-seven (35%) of the 220 species in our dataset are threatened based on the IUCN Red List; 12 as Critical Endangered (CR), 30 Endangered (EN), and 35 Vulnerable (VU). These salamanders should receive special attention, especially if they occupy small ranges and have low genetic diversity. Limitations and challenges One major limitation for macrogenetic studies is the uneven spatial sampling of available sequences, which can lead to an incomplete picture of the distribution of genetic diversity. There are also sampling biases among taxa, such that some species are well characterized while others are omitted from global comparisons. Another limitation is the use of a single mitochondrial gene to calculate genetic diversity and genetic distances. Although mtDNA is widely used in phylogenetic and population genetic studies because it evolves quickly ( Galtier et al., 2009 ), it provides an incomplete picture of intraspecific diversity because it is a single locus, haploid, and solely maternally inherited ( Moore, 1995 ). Moreover, although there is a potential association between lineage age and genetic diversity, it is important to note that divergence times estimated from different loci can have distinct coalescence histories and are often younger than species ages. This suggests that a relationship between divergence times across loci and genetic variation inferred from a single locus (mtDNA) is possible but not necessarily straightforward ( Pie and Caron, 2023 ). However, single locus macrogenetic studies are still valuable because the amount and availability of data elucidate patterns associated with historical processes and variability within species. Furthermore, the large genome size of salamanders hampers the use of nuclear genomic data in studies on this scale. These limitations reinforce the need for additional geographic sampling of high-quality tissues that include broader taxonomic coverage and facilitate new genomic sequencing efforts. The most important challenge for our study was the time required to retrieve the geographic coordinates associated with each Cytb sequence. Most sequences in open access databases such as GenBank do not have their latitude and longitude associated with them, which limits studies of spatial genetic variation. Efforts such as phylogatR ( Pelletier et al. 2022 ) aggregate genetic sequences associated with georeferenced occurrences from global databases using automated pipelines, but researchers providing the sequences must include the proper metadata when uploading their data for this approach to work. We encourage scientists to share spatial data and museum catalogue numbers when applicable along with sequences when they are published, ensuring data reproducibility and extendibility. Conclusions Our study underscores the value of integrating spatial information into macrogenetic studies. Our results support the hypotheses that variables associated with population size are positively correlated with genetic variation and that salamanders tend to exhibit spatial patterns of genetic variation that also differ across biogeographic realms. Based on our results, we reject the hypothesis that traits associated with reproductive strategies predict mitochondrial genetic variation in salamanders; and we partially support the hypothesis that older species tend to accumulate higher generic variation, as explained above. Our results suggest the importance of range-wide climatic variation, local environmental heterogeneity, and historical demographic events for influencing genetic variation. Future amphibian macrogenetic and genomic studies can incorporate more explicit tests of range expansion and contraction to better understand these processes. Finally, our findings have broad implications for establishing conservation priorities and highlighting regions where sampling has been historically scarce, particularly in biodiversity hotspots that warrant further attention. Data and code availability statement The data and code (e.g., R scripts) that support the results of this research are openly available at https://github.com/luchoamador/amphibian_macrogenetics . Large folders and other supplementary materials are available in Dryad https://doi.org/10.5061/dryad.51c59zwjx . Funder Information Declared National Science Foundation , DEB-2112946 , DBI-1911293 Footnotes The main changes incorporated in this version are: We include species age (divergence time) as a predictor of genetic variation, based on the Stewart and Wiens (2025) study ( https://doi.org/10.1016/j.ympev.2024.108272 ). We discuss the geological events that could have influenced the genetic variation of the salamander species. We have included the statement –Data and Code Availability Statement-in the main text. We have carefully checked the alignments and species distribution ranges for all species and we revised the analyses based on the new findings. For example, our dataset increased from 214 species to 220 species. We modified the title following the new results. The new title is: Caudata macrogenetics: Geographic attributes and lineage age predict global patterns of mitochondrial genetic variation in salamanders. https://github.com/luchoamador/amphibian_macrogenetics https://doi.org/10.5061/dryad.51c59zwjx References 1. ↵ Alzate , A. , Rozzi , R. , Velasco , J. A. , Robertson , D. R. , Zizka , A. , Tobias , J. A. , Hill , A. , Bacon , C. D. , Janzen , T. , Pellissier , L. , Van Der Plas , F. , Rosindell , J. , & Onstein , R. E. ( 2025 ). Evolutionary age correlates with range size across plants and animals . Nature Communications , 16 ( 1 ), 7894 . doi: 10.1038/s41467-025-62124-y OpenUrl CrossRef PubMed 2. ↵ Amador , L. , Arroyo Torres , I. , & Barrow , L. N . ( 2024 ). Machine learning and phylogenetic models identify predictors of genetic variation in Neotropical amphibians . Journal of Biogeography , 51 ( 5 ), 909 – 923 . doi: 10.1111/jbi.14795 OpenUrl CrossRef 3. ↵ AmphibiaWeb . ( 2024 ). University of California, Berkeley, CA, USA. Accessed 3 October 2024 . 4. ↵ Antonelli , A . ( 2022 ). The rise and fall of Neotropical biodiversity . Botanical Journal of the Linnean Society , 199 ( 1 ), 8 – 24 . doi: 10.1093/botlinnean/boab061 OpenUrl CrossRef 5. ↵ Auteri , G. G. , Marchán-Rivadeneira , M. R. , Olson , D. H. , & Knowles , L. L . ( 2022 ). Landscape connectivity among coastal giant salamander ( Dicamptodon tenebrosus ) populations shows no association with land use, fire frequency, or river drainage but exhibits genetic signatures of potential conservation concern . PLOS ONE , 17 ( 6 ), e0268882 . doi: 10.1371/journal.pone.0268882 OpenUrl CrossRef PubMed 6. ↵ Bani , L. , Pisa , G. , Luppi , M. , Spilotros , G. , Fabbri , E. , Randi , E. , & Orioli , V . ( 2015 ). Ecological connectivity assessment in a strongly structured fire salamander ( Salamandra salamandra ) population . Ecology and Evolution , 5 ( 16 ), 3472 – 3485 . doi: 10.1002/ece3.1617 OpenUrl CrossRef 7. ↵ Barrow , L. N. , Masiero Da Fonseca , E. , Thompson , C. E. P. , & Carstens , B. C. ( 2021 ). Predicting amphibian intraspecific diversity with machine learning: Challenges and prospects for integrating traits, geography, and genetic data . Molecular Ecology Resources , 21 ( 8 ), 2818 – 2831 . doi: 10.1111/1755-0998.13303 OpenUrl CrossRef 8. ↵ Bivand R , Pebesma E , Gómez-Rubio V . ( 2013 ). Applied spatial data analysis with R, Second edition . Springer , NY . https://asdar-book.org/ . 9. ↵ Blanchet , S. , Prunier , J. G. , & De Kort , H. ( 2017 ). Time to Go Bigger: Emerging Patterns in Macrogenetics . Trends in Genetics , 33 ( 9 ), 579 – 580 . doi: 10.1016/j.tig.2017.06.007 OpenUrl CrossRef PubMed 10. ↵ Blomberg , S. P. , Garland , T. , & Ives , A. R . ( 2003 ). Testing for phylogenetic signal in comparative data: behavioral traits are more labile . Evolution , 57 ( 4 ), 717 – 745 . OpenUrl CrossRef PubMed Web of Science 11. ↵ Bonato , L. , Corbetta , A. , Giovine , G. , Romanazzi , E. , Šunje , E. , Vernesi , C. , & Crestanello , B . ( 2018 ). Diversity among peripheral populations: Genetic and evolutionary differentiation of Salamandra atra at the southern edge of the Alps . Journal of Zoological Systematics and Evolutionary Research , 56 ( 4 ), 533 – 548 . doi: 10.1111/jzs.12224 OpenUrl CrossRef 12. ↵ Bonferroni , C. E . ( 1936 ) Teoria statistica delle classi e calcolo delle probabilita . Pubblicazioni del R Istituto Superiore di Scienze Economiche e Commerciali di Firenze 8 , 3 – 62 , 1936. OpenUrl CrossRef 13. ↵ Calenge C ( 2024 ). adehabitatHR: Home Range Estimation . R package version 0.4.22 , . 14. ↵ Cambon , J. , Hernangómez , D. , Belanger , C. , & Possenriede , D . ( 2021 ). tidygeocoder: An R package for geocoding . Journal of Open Source Software , 6 ( 65 ), 3544 . doi: 10.21105/joss.03544 OpenUrl CrossRef 15. ↵ Carnaval , A. C. , Hickerson , M. J. , Haddad , C. F. B. , Rodrigues , M. T. , & Moritz , C . ( 2009 ). Stability Predicts Genetic Diversity in the Brazilian Atlantic Forest Hotspot . Science , 323 ( 5915 ), 785 – 789 . doi: 10.1126/science.1166955 OpenUrl Abstract / FREE Full Text 16. ↵ Corbett-Detig , R. B. , Hartl , D. L. , & Sackton , T. B . ( 2015 ). Natural Selection Constrains Neutral Diversity across A Wide Range of Species . PLOS Biology , 13 ( 4 ), e1002112 . doi: 10.1371/journal.pbio.1002112 OpenUrl CrossRef PubMed 17. ↵ Crispo , E. , & Hendry , A. P . ( 2005 ). Does time since colonization influence isolation by distance? A meta-analysis. Conservation Genetics , 6 ( 5 ), 665 – 682 . doi: 10.1007/s10592-005-9026-4 OpenUrl CrossRef 18. ↵ De Kort , H. , Prunier , J. G. , Ducatez , S. , Honnay , O. , Baguette , M. , Stevens , V. M. , & Blanchet , S. ( 2021 ). Life history, climate and biogeography interactively affect worldwide genetic diversity of plant and animal populations . Nature Communications , 12 ( 1 ), 516 . doi: 10.1038/s41467-021-20958-2 OpenUrl CrossRef PubMed 19. ↵ Denoël , M. , Joly , P. , & Whiteman , H. H . ( 2005 ). Evolutionary ecology of facultative paedomorphosis in newts and salamanders . Biological Reviews , 80 ( 4 ), 663 – 671 . doi: 10.1017/S1464793105006858 OpenUrl CrossRef PubMed 20. ↵ Dunnington , D. ( 2023 ). ggspatial: Spatial Data Framework for ggplot2. R package version 1.1.9 , . 21. ↵ Eckert , C. G. , Samis , K. E. , & Lougheed , S. C . ( 2008 ). Genetic variation across species’ geographical ranges: The central–marginal hypothesis and beyond . Molecular Ecology , 17 ( 5 ), 1170 – 1188 . doi: 10.1111/j.1365-294X.2007.03659.x OpenUrl CrossRef PubMed Web of Science 22. ↵ Edgar , R. C . ( 2004 ). MUSCLE: a multiple sequence alignment method with reduced time and space complexity . BMC Bioinformatics , 5 ( 1 ), 113 . doi: 10.1186/1471-2105-5-113 OpenUrl CrossRef PubMed 23. ↵ Ellegren , H. , & Galtier , N . ( 2016 ). Determinants of genetic diversity . Nature Reviews Genetics , 17 ( 7 ), 422 – 433 . doi: 10.1038/nrg.2016.58 OpenUrl CrossRef PubMed 24. ↵ Escobar , S. , Helmstetter , A. J. , Jarvie , S. , Montúfar , R. , Balslev , H. , & Couvreur , T. L. P . ( 2021 ). Pleistocene climatic fluctuations promoted alternative evolutionary histories in Phytelephas aequatorialis , an endemic palm from western Ecuador . Journal of Biogeography , 48 ( 5 ), 1023 – 1037 . doi: 10.1111/jbi.14055 OpenUrl CrossRef 25. ↵ Felsenstein , J . ( 1985 ). Phylogenies and the Comparative Method . The American Naturalist , 125 ( 1 ), 1 – 15 . OpenUrl CrossRef Web of Science 26. ↵ Fick , S.E. and R.J. Hijmans , ( 2017 ). WorldClim 2: new 1km spatial resolution climate surfaces for global land areas . International Journal of Climatology 37 ( 12 ): 4302 – 4315 . OpenUrl CrossRef 27. ↵ Fonseca , E. M. , Garda , A. A. , Oliveira , E. F. , Camurugi , F. , Magalhães , F. D. M. , Lanna , F. M. , Zurano , J. P. , Marques , R. , Vences , M. , & Gehara , M . ( 2021 ). The riverine thruway hypothesis: Rivers as a key mediator of gene flow for the aquatic paradoxical frog Pseudis tocantins (Anura, Hylidae) . Landscape Ecology , 36 ( 10 ), 3049 – 3060 . doi: 10.1007/s10980-021-01257-z OpenUrl CrossRef 28. ↵ Fonseca , E. M. , Pelletier , T. A. , Decker , S. K. , Parsons , D. J. , & Carstens , B. C . ( 2023 ). Pleistocene glaciations caused the latitudinal gradient of within-species genetic diversity . Evolution Letters , 7 ( 5 ), 331 – 338 . doi: 10.1093/evlett/qrad030 OpenUrl CrossRef PubMed 29. ↵ Frankham , R . ( 1996 ). Relationship of Genetic Variation to Population Size in Wildlife . Conservation Biology , 10 ( 6 ), 1500 – 1508 . doi: 10.1046/j.1523-1739.1996.10061500.x OpenUrl CrossRef Web of Science 30. ↵ French , C. M. , Bertola , L. D. , Carnaval , A. C. , Economo , E. P. , Kass , J. M. , Lohman , D. J. , Marske , K. A. , Meier , R. , Overcast , I. , Rominger , A. J. , Staniczenko , P. P. A. , & Hickerson , M. J . ( 2023 ). Global determinants of insect mitochondrial genetic diversity . Nature Communications , 14 ( 1 ), 5276 . doi: 10.1038/s41467-023-40936-0 OpenUrl CrossRef PubMed 31. ↵ Fu , J. , & Wen , L . ( 2023 ). Impacts of Quaternary glaciation, geological history and geography on animal species history in continental East Asia: A phylogeographic review . Molecular Ecology , 32 ( 16 ), 4497 – 4514 . doi: 10.1111/mec.17053 OpenUrl CrossRef 32. ↵ Fusco , N. A. , Pehek , E. , & Munshi South , J . ( 2021 ). Urbanization reduces gene flow but not genetic diversity of stream salamander populations in the New York City metropolitan area . Evolutionary Applications , 14 ( 1 ), 99 – 116 . doi: 10.1111/eva.13025 OpenUrl CrossRef 33. ↵ Galtier , N. , Nabholz , B. , Glémin , S. , & Hurst , G. D. D . ( 2009 ). Mitochondrial DNA as a marker of molecular diversity: A reappraisal . Molecular Ecology , 18 ( 22 ), 4541 – 4550 . doi: 10.1111/j.1365-294X.2009.04380.x OpenUrl CrossRef PubMed Web of Science 34. ↵ Gillman , L. N. , & Wright , S. D . ( 2014 ). Species richness and evolutionary speed: The influence of temperature, water and area . Journal of Biogeography , 41 ( 1 ), 39 – 51 . doi: 10.1111/jbi.12173 OpenUrl CrossRef Web of Science 35. ↵ Hanson , J. O. , Rhodes , J. R. , Riginos , C. , & Fuller , R. A . ( 2017 ). Environmental and geographic variables are effective surrogates for genetic variation in conservation planning . Proceedings of the National Academy of Sciences , 114 ( 48 ), 12755 – 12760 . doi: 10.1073/pnas.1711009114 OpenUrl Abstract / FREE Full Text 36. ↵ Hijmans R ( 2022 ). geosphere: Spherical Trigonometry . R package version 1 . 5 – 18 , . OpenUrl 37. ↵ Hijmans RJ , Barbosa M , Ghosh A , Mandel A ( 2024 ). geodata: Download Geographic Data . R package version 0 . 6 – 2 , . OpenUrl 38. ↵ Hollister J , Shah T , Nowosad J , Robitaille A , Beck M , Johnson M ( 2025 ). elevatr: Access Elevation Data from Various APIs . doi: 10.5281/zenodo.8335450 , R package version 0.99.1 , https://github.com/usepa/elevatr/ . OpenUrl CrossRef 39. ↵ Iannella , M. , Cerasoli , F. , Lunghi , E. , Console , G. , Biondi , M. , & Sillero , N . ( 2025 ). Climate Change Effects on the Only Western Palearctic Plethodontids: Range Changes and Possible Depletion of Intraspecific Genetic Diversity . Journal of Biogeography , 52 ( 3 ), 686 – 698 . doi: 10.1111/jbi.15064 OpenUrl CrossRef 40. ↵ Ives , A. R. , & Helmus , M. R . ( 2011 ). Generalized linear mixed models for phylogenetic analyses of community structure . Ecological Monographs , 81 ( 3 ), 511 – 525 . doi: 10.1890/10-1264.1 OpenUrl CrossRef Web of Science 41. ↵ Janzen , D. H . ( 1967 ). Why Mountain Passes are Higher in the Tropics . The American Naturalist , 101 ( 919 ), 233 – 249 . doi: 10.1086/282487 OpenUrl CrossRef Web of Science 42. ↵ Jenkins , D. G. , Carey , M. , Czerniewska , J. , Fletcher , J. , Hether , T. , Jones , A. , Knight , S. , Knox , J. , Long , T. , Mannino , M. , McGuire , M. , Riffle , A. , Segelsky , S. , Shappell , L. , Sterner , A. , Strickler , T. , & Tursi , R . ( 2010 ). A meta analysis of isolation by distance: Relic or reference standard for landscape genetics? Ecography , 33 ( 2 ), 315 – 320 . doi: 10.1111/j.1600-0587.2010.06285.x OpenUrl CrossRef 43. ↵ Jetz , W. , & Pyron , R. A . ( 2018 ). The interplay of past diversification and evolutionary isolation with present imperilment across the amphibian tree of life . Nature Ecology & Evolution , 2 ( 5 ), 850 – 858 . doi: 10.1038/s41559-018-0515-5 OpenUrl CrossRef PubMed 44. ↵ Kardos , M. , Armstrong , E. E. , Fitzpatrick , S. W. , Hauser , S. , Hedrick , P. W. , Miller , J. M. , Tallmon , D. A. , & Funk , W. C . ( 2021 ). The crucial role of genome-wide genetic variation in conservation . Proceedings of the National Academy of Sciences , 118 ( 48 ), e2104642118 . doi: 10.1073/pnas.2104642118 OpenUrl Abstract / FREE Full Text 45. ↵ Kimura , M . ( 1968 ). Evolutionary rate at the molecular level . Nature , 217 ( 5129 ), 624 – 626 . doi: 10.1038/217624a0 OpenUrl CrossRef PubMed Web of Science 46. ↵ Klingenberg , C. P. , & Gidaszewski , N. A . ( 2010 ). Testing and Quantifying Phylogenetic Signals and Homoplasy in Morphometric Data . Systematic Biology , 59 ( 3 ), 245 – 261 . doi: 10.1093/sysbio/syp106 OpenUrl CrossRef PubMed 47. ↵ Kozak , K. H. , & Wiens , J. J . ( 2006 ). Does niche conservatism promote speciation? A case study in North American salamanders . Evolution , 60 ( 12 ), 2604 – 2621 . doi: 10.1111/j.0014-3820.2006.tb01893.x OpenUrl CrossRef PubMed Web of Science 48. ↵ Kuhn , M . ( 2008 ). “ Building Predictive Models in R Using the caret Package .” Journal of Statistical Software , 28 ( 5 ), 1 – 26 . doi: 10.18637/jss.v028.i05 , https://www.jstatsoft.org/index.php/jss/article/view/v028i05 . OpenUrl CrossRef PubMed 49. ↵ Lacy , R. C . ( 1987 ). Loss of Genetic Diversity from Managed Populations: Interacting Effects of Drift, Mutation, Immigration, Selection, and Population Subdivision . Conservation Biology , 1 ( 2 ), 143 – 158 . OpenUrl CrossRef 50. ↵ Lai , J. , He , Y. , Hou , M. , Zhang , A. , Wang , G. , & Mao , L . ( 2025 ). Evaluating the relative importance of phylogeny and predictors in phylogenetic generalized linear models using the phylolm.hp R package . Plant Diversity , 47 ( 5 ), 709 – 717 . doi: 10.1016/j.pld.2025.06.003 OpenUrl CrossRef PubMed 51. ↵ Larsson , A. ( 2014 ). AliView: A fast and lightweight alignment viewer and editor for large datasets . Bioinformatics , 30 ( 22 ), 3276 – 3278 . doi: 10.1093/bioinformatics/btu531 OpenUrl CrossRef PubMed 52. ↵ Lawrence , E. R. , Pedersen , E. J. , & Fraser , D. J . ( 2023 ). Macrogenetics reveals multifaceted influences of environmental variation on vertebrate population genetic diversity across the Americas . Molecular Ecology , 32 ( 16 ), 4557 – 4569 . doi: 10.1111/mec.17059 OpenUrl CrossRef 53. Lee Yaw , J. A. , & Irwin , D. E. . ( 2012 ). Large geographic range size reflects a patchwork of divergent lineages in the long toed salamander ( Ambystoma macrodactylum ) . Journal of Evolutionary Biology , 25 ( 11 ), 2276 – 2287 . doi: 10.1111/j.1420-9101.2012.02604.x OpenUrl CrossRef PubMed 54. ↵ Leigh , D. M. , Van Rees , C. B. , Millette , K. L. , Breed , M. F. , Schmidt , C. , Bertola , L. D. , Hand , B. K. , Hunter , M. E. , Jensen , E. L. , Kershaw , F. , Liggins , L. , Luikart , G. , Manel , S. , Mergeay , J. , Miller , J. M. , Segelbacher , G. , Hoban , S. , & Paz-Vinas , I. ( 2021 ). Opportunities and challenges of macrogenetic studies . Nature Reviews Genetics , 22 ( 12 ), 791 – 807 . doi: 10.1038/s41576-021-00394-0 OpenUrl CrossRef PubMed 55. ↵ Li , D. , Dinnage , R. , Nell , L. A. , Helmus , M. R. , & Ives , A. R . ( 2020 ). phyr: An r package for phylogenetic species distribution modelling in ecological communities . Methods in Ecology and Evolution , 11 ( 11 ), 1455 – 1463 . doi: 10.1111/2041-210X.13471 OpenUrl CrossRef 56. ↵ Liaw , A. , Wiener , M . ( 2002 ). Classification and Regression by randomForest . R News , 2 ( 3 ), 18 – 22 . . OpenUrl CrossRef 57. López Delgado , J. , & Meirmans , P. G. ( 2022 ). History or demography? Determining the drivers of genetic variation in North American plants . Molecular Ecology , 31 ( 7 ), 1951 – 1962 . doi: 10.1111/mec.16230 OpenUrl CrossRef 58. ↵ Lowe , W. H. , Kovach , R. P. , & Allendorf , F. W . ( 2017 ). Population Genetics and Demography Unite Ecology and Evolution . Trends in Ecology & Evolution , 32 ( 2 ), 141 – 152 . doi: 10.1016/j.tree.2016.12.002 OpenUrl CrossRef PubMed 59. ↵ Luedtke , J. A. , Chanson , J. , Neam , K. , Hobin , L. , Maciel , A. O. , Catenazzi , A. , Borzée , A. , Hamidy , A. , Aowphol , A. , Jean , A. , Sosa-Bartuano , Á. , Fong G. A. , De Silva A. , Fouquet A. , Angulo A. , Kidov A. A. , Muñoz Saravia A. , Diesmos A. C. , Tominaga A. , … Stuart , S. N. ( 2023 ). Ongoing declines for the world’s amphibians in the face of emerging threats . Nature , 622 ( 7982 ), 308 – 314 . doi: 10.1038/s41586-023-06578-4 OpenUrl CrossRef 60. Malcher , G. , Amorim , A.L. , Ferreira , P. et al. ( 2023 ). First evaluation of the population genetics and aspects of the evolutionary history of the Amazonian snook, Centropomus irae , and its association with the Amazon plume . Hydrobiologia 850 , 2115 – 2125 . doi: 10.1007/s10750-023-05223-5 OpenUrl CrossRef 61. ↵ Mayr , E . ( 2001 ). What Evolution Is . New York, NY : Basic Books . 62. ↵ McPeek , M. A. , & Brown , J. M . ( 2007 ). Clade Age and Not Diversification Rate Explains Species Richness among Animal Taxa . The American Naturalist , 169 ( 4 ), E97 – E106 . doi: 10.1086/512135 OpenUrl CrossRef PubMed Web of Science 63. ↵ Meng , H. , Li , X. , & Qiao , P . ( 2014 ). Population Structure, Historical Biogeography and Demographic History of the Alpine Toad Scutiger ningshanensis in the Tsinling Mountains of Central China . PLoS ONE , 9 ( 6 ), e100729 . doi: 10.1371/journal.pone.0100729 OpenUrl CrossRef PubMed 64. ↵ Millette , K. L. , Fugère , V. , Debyser , C. , Greiner , A. , Chain , F. J. J. , & Gonzalez , A . ( 2021 ). Refining analyses of existing data sets is valuable for macrogenetics: A response to Paz Vinas , Jensen et al. , (2021). Ecology Letters , 24 (6), 1285–1286. doi: 10.1111/ele.13733 OpenUrl CrossRef 65. ↵ Liedtke , H.C. , Wiens , J.J. & Gomez-Mestre , I . ( 2022 ). The evolution of reproductive modes and life cycles in amphibians . Nat Commun 13 , 7039 . doi: 10.1038/s41467-022-34474-4 OpenUrl CrossRef PubMed 66. ↵ Miraldo , A. , Li , S. , Borregaard , M. K. , Flórez-Rodríguez , A. , Gopalakrishnan , S. , Rizvanovic , M. , Wang , Z. , Rahbek , C. , Marske , K. A. , & Nogués-Bravo , D . ( 2016 ). An Anthropocene map of genetic diversity . Science , 353 ( 6307 ), 1532 – 1535 . doi: 10.1126/science.aaf4381 OpenUrl Abstract / FREE Full Text 67. ↵ F. E. Zachos Mittermeier , R. A. , Turner , W. R. , Larsen , F. W. , Brooks , T. M. , & Gascon , C. ( 2011 ). Global Biodiversity Conservation: The Critical Role of Hotspots . In F. E. Zachos & J. C. Habel (Eds.), Biodiversity Hotspots (pp. 3 – 22 ). Springer Berlin Heidelberg . doi: 10.1007/978-3-642-20992-5_1 OpenUrl CrossRef 68. ↵ Moore , W. S . ( 1995 ). Inferring phylogenies from mtDNA variation: mitochondrial-gene trees versus nuclear-gene trees . Evolution , 49 ( 4 ), 718 – 726 . OpenUrl CrossRef PubMed Web of Science 69. ↵ Oksanen J , Simpson G , Blanchet F , Kindt R , Legendre P , Minchin P , O’Hara R , Solymos P , Stevens M , Szoecs E , Wagner H , Barbour M , Bedward M , Bolker B , Borcard D , Carvalho G , Chirico M , De Caceres M , Durand S , Evangelista H , FitzJohn R , Friendly M , Furneaux B , Hannigan G , Hill M , Lahti L , McGlinn D , Ouellette M , Ribeiro Cunha E , Smith T , Stier A , Ter Braak C , Weedon J ( 2024 ). vegan: Community Ecology Package . R package version 2 . 6 – 8 , OpenUrl 70. ↵ Oliveira , B. F. , São-Pedro , V. A. , Santos-Barrera , G. , Penone , C. , & Costa , G. C . ( 2017 ). AmphiBIO, a global database for amphibian ecological traits . Scientific Data , 4 ( 1 ), 170123 . doi: 10.1038/sdata.2017.123 OpenUrl CrossRef PubMed 71. ↵ Pagel , M . ( 1999 ). Inferring the historical patterns of biological evolution . Nature , 401 ( 6756 ), 877 – 884 . doi: 10.1038/44766 OpenUrl CrossRef PubMed Web of Science 72. ↵ Pan , T. , Wang , H. , Orozcoterwengel , P. , Hu , C.-C. , Wu , G.-Y. , Qian , L.-F. , Sun , Z.-L. , Shi , W.-B. , Yan , P. , Wu , X.-B. , & Zhang , B.-W . ( 2019 ). Long-term sky islands generate highly divergent lineages of a narrowly distributed stream salamander (Pachyhynobius shangchengensis) in mid-latitude mountains of East Asia . BMC Evolutionary Biology , 19 ( 1 ), 1 . doi: 10.1186/s12862-018-1333-8 OpenUrl CrossRef PubMed 73. ↵ Paradis , E . ( 2010 ). pegas: An R package for population genetics with an integrated–modular approach . Bioinformatics , 26 ( 3 ), 419 – 420 . doi: 10.1093/bioinformatics/btp696 OpenUrl CrossRef PubMed Web of Science 74. ↵ Paradis , E. , & Schliep , K. ( 2019 ). ape 5.0: An environment for modern phylogenetics and evolutionary analyses in R . Bioinformatics , 35 ( 3 ), 526 – 528 . doi: 10.1093/bioinformatics/bty633 OpenUrl CrossRef PubMed 75. ↵ Parsons , D. J. , Pelletier , T. A. , Wieringa , J. G. , Duckett , D. J. , & Carstens , B. C . ( 2022 ). Analysis of biodiversity data suggests that mammal species are hidden in predictable places . Proceedings of the National Academy of Sciences , 119 ( 14 ), e2103400119 . doi: 10.1073/pnas.2103400119 OpenUrl CrossRef PubMed 76. ↵ Parsons , D. J. , Green , A. E. , Carstens , B. C. , & Pelletier , T. A . ( 2024 ). Predicting genetic biodiversity in salamanders using geographic, climatic, and life history traits . PLOS ONE , 19 ( 10 ), e0310932 . doi: 10.1371/journal.pone.0310932 OpenUrl CrossRef PubMed 77. ↵ Park , K. Y. , Lucas , M. , Chaulk , A. , Matter , S. F. , Roland , J. , & Keyghobadi , N . ( 2024 ). Immigration allows population persistence and maintains genetic diversity despite an attempted experimental extinction . Royal Society Open Science , 11 ( 7 ), 240557 . doi: 10.1098/rsos.240557 OpenUrl CrossRef PubMed 78. ↵ Pauls , S. U. , Nowak , C. , Bálint , M. , & Pfenninger , M . ( 2013 ). The impact of global climate change on genetic diversity within populations and species . Molecular Ecology , 22 ( 4 ), 925 – 946 . doi: 10.1111/mec.12152 OpenUrl CrossRef 79. ↵ Paz , A. , Ibáñez , R. , Lips , K.R. and Crawford , A.J . ( 2015 ), Testing the role of ecology and life history in structuring genetic variation across a landscape: a trait-based phylogeographic approach . Molecular Ecology , 24 : 3723 – 3737 . https://doi-org.libproxy.unm.edu/10.1111/mec.13275 OpenUrl CrossRef 80. Paz Vinas , I. , Jensen , E. L. , Bertola , L. D. , Breed , M. F. , Hand , B. K. , Hunter , M. E. , Kershaw , F. , Leigh , D. M. , Luikart , G. , Mergeay , J. , Miller , J. M. , Van Rees , C. B. , Segelbacher , G. , & Hoban , S. ( 2021 ). Macrogenetic studies must not ignore limitations of genetic markers and scale . Ecology Letters , 24 ( 6 ), 1282 – 1284 . doi: 10.1111/ele.13732 OpenUrl CrossRef PubMed 81. ↵ Pebesma , E ., 2018 . Simple Features for R: Standardized Support for Spatial Vector Data . The R Journal 10 ( 1 ), 439 – 446 , doi: 10.32614/RJ-2018-009 OpenUrl CrossRef 82. ↵ Pebesma , E. , & Bivand , R . ( 2023 ). Spatial Data Science: With Applications in R . Chapman and Hall/CRC . doi: 10.1201/9780429459016 OpenUrl CrossRef 83. ↵ Pelletier , T. A. , & Carstens , B. C . ( 2018 ). Geographical range size and latitude predict population genetic structure in a global survey . Biology Letters , 14 ( 1 ), 20170566 . doi: 10.1098/rsbl.2017.0566 OpenUrl CrossRef PubMed 84. ↵ Pelletier , T. A. , Parsons , D. J. , Decker , S. K. , Crouch , S. , Franz , E. , Ohrstrom , J. , & Carstens , B. C . ( 2022 ). phylogatR: Phylogeographic data aggregation and repurposing . Molecular Ecology Resources , 22 ( 8 ), 2830 – 2842 . doi: 10.1111/1755-0998.13673 OpenUrl CrossRef PubMed 85. Petit Marty , N. , Vázquez Luis , M. , & Hendriks , I. E. ( 2021 ). Use of the nucleotide diversity in COI mitochondrial gene as an early diagnostic of conservation status of animal species . Conservation Letters , 14 ( 1 ), e12756 . doi: 10.1111/conl.12756 OpenUrl CrossRef 86. ↵ Pie , M. R. , & Caron , F. S . ( 2023 ). Substantial variation in species ages among vertebrate clades . Evolutionary Journal of the Linnean Society , 2 ( 1 ), kzad006 . doi: 10.1093/evolinnean/kzad006 OpenUrl CrossRef 87. ↵ R Core Team ( 2023 ). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing , Vienna, Austria. . 88. ↵ Ramírez Barahona , S. , & Eguiarte , L. E. ( 2013 ). The role of glacial cycles in promoting genetic diversity in the Neotropics: The case of cloud forests during the Last Glacial Maximum . Ecology and Evolution , 3 ( 3 ), 725 – 738 . doi: 10.1002/ece3.483 OpenUrl CrossRef PubMed 89. ↵ Reed , D. H. , & Frankham , R . ( 2003 ). Correlation between Fitness and Genetic Diversity . Conservation Biology , 17 ( 1 ), 230 – 237 . doi: 10.1046/j.1523-1739.2003.01236.x OpenUrl CrossRef 90. ↵ Revell , L. J . ( 2012 ). phytools: An R package for phylogenetic comparative biology (and other things): phytools: R package . Methods in Ecology and Evolution , 3 ( 2 ), 217 – 223 . doi: 10.1111/j.2041-210X.2011.00169.x OpenUrl CrossRef PubMed 91. ↵ Riberon , A. , Sotiriou , E. , Miaud , C. , Andreone , F. , & Taberlet , P . ( 2002 ). Lack of genetic diversity in Salamandra lanzai revealed by Cytochrome b gene sequences . Copeia , 2002 ( 1 ), 229 – 232 . OpenUrl CrossRef 92. ↵ Rios , N. E. & Bart , H. L . ( 2010 ). GEOLocate (Version 3.22). [Computer software]. Belle Chasse , LA : Tulane University Museum of Natural History . 93. ↵ Román Palacios , C. ( 2023 ). The phruta r package: Increasing access, reproducibility and transparency in phylogenetic analyses . Methods in Ecology and Evolution , 14 ( 9 ), 2284 – 2299 . doi: 10.1111/2041-210X.14147 OpenUrl CrossRef 94. ↵ Rovito , S. M . ( 2017 ). The Geography of Speciation in Neotropical Salamanders . Herpetologica , 73 ( 3 ), 229 – 241 . doi: 10.1655/HERPETOLOGICA-D-16-00077.1 OpenUrl CrossRef 95. ↵ Rovito , S. M. , & Schoville , S. D . ( 2017 ). Testing models of refugial isolation, colonization and population connectivity in two species of montane salamanders . Heredity , 119 ( 4 ), 265 – 274 . doi: 10.1038/hdy.2017.31 OpenUrl CrossRef PubMed 96. ↵ RS-eco ( 2023 ). rasterSp: R Package to rasterize and summarise IUCN range maps. R package version 0.0.1 , . 97. ↵ Schierenbeck , K. A . ( 2017 ). Population-level genetic variation and climate change in a biodiversity hotspot . Annals of Botany , 119 ( 2 ), 215 – 228 . doi: 10.1093/aob/mcw214 OpenUrl CrossRef PubMed 98. ↵ Schmitt , T . ( 2007 ). Molecular biogeography of Europe: Pleistocene cycles and postglacial trends . Frontiers in Zoology , 4 ( 1 ), 11 . doi: 10.1186/1742-9994-4-11 OpenUrl CrossRef PubMed 99. ↵ Scholl , J. P. , & Wiens , J. J . ( 2016 ). Diversification rates and species richness across the Tree of Life . Proceedings of the Royal Society B: Biological Sciences , 283 ( 1838 ), 20161334 . doi: 10.1098/rspb.2016.1334 OpenUrl CrossRef PubMed 100. Segovia Ramírez , M. G. , Ramírez Sánchez , O. , Decena Segarra , L. P. , Rios Carlos , H. , & Rovito , S. M. ( 2023 ). Determinants of genetic diversity in Neotropical salamanders (Plethodontidae: Bolitoglossini) . Ecology and Evolution , 13 ( 11 ), e10707 . doi: 10.1002/ece3.10707 OpenUrl CrossRef 101. ↵ Sexton , J. P. , Hangartner , S. B. , & Hoffmann , A. A . ( 2014 ). Genetic isolation by environment or distance: which pattern of gene flow is most common? Special Section . Evolution , 68 ( 1 ), 1 – 15 . doi: 10.1111/evo.12258 OpenUrl CrossRef PubMed Web of Science 102. ↵ Sexton , J. P. , Clemens , M. , Bell , N. , Hall , J. , Fyfe , V. , & Hoffmann , A. A . ( 2024 ). Patterns and effects of gene flow on adaptation across spatial scales: Implications for management . Journal of Evolutionary Biology , 37 ( 6 ), 732 – 745 . doi: 10.1093/jeb/voae064 OpenUrl CrossRef PubMed 103. ↵ Shafer , A. B. A. , Cullingham , C. I. , Côté , S. D. , & Coltman , D. W . ( 2010 ). Of glaciers and refugia: A decade of study sheds new light on the phylogeography of northwestern North America . Molecular Ecology , 19 ( 21 ), 4589 – 4621 . doi: 10.1111/j.1365-294X.2010.04828.x OpenUrl CrossRef PubMed Web of Science 104. ↵ Smith , B. T. , Seeholzer , G. F. , Harvey , M. G. , Cuervo , A. M. , & Brumfield , R. T . ( 2017 ). A latitudinal phylogeographic diversity gradient in birds . PLOS Biology , 15 ( 4 ), e2001073 . doi: 10.1371/journal.pbio.2001073 OpenUrl CrossRef PubMed 105. ↵ Stevens , G. C . ( 1989 ). The Latitudinal Gradient in Geographical Range: How so Many Species Coexist in the Tropics . The American Naturalist , 133 ( 2 ), 240 – 256 . doi: 10.1086/284913 OpenUrl CrossRef Web of Science 106. ↵ Stewart , A. A. , & Wiens , J. J . ( 2025 ). A time-calibrated salamander phylogeny including 765 species and 503 genes . Molecular Phylogenetics and Evolution , 204 , 108272 . doi: 10.1016/j.ympev.2024.108272 OpenUrl CrossRef PubMed 107. ↵ Tajima , F . ( 1989 ). Statistical methods to test for nucleotide mutation hypothesis by DNA polymorphism . Genetics 123 : 585 – 595 . OpenUrl Abstract / FREE Full Text 108. ↵ Theodoridis , S. , Fordham , D. A. , Brown , S. C. , Li , S. , Rahbek , C. , & Nogues-Bravo , D . ( 2020 ). Evolutionary history and past climate change shape the distribution of genetic diversity in terrestrial mammals . Nature Communications , 11 ( 1 ), 2557 . doi: 10.1038/s41467-020-16449-5 OpenUrl CrossRef 109. ↵ Tingley , R. , & Dubey , S . ( 2012 ). Disparity in the timing of vertebrate diversification events between the northern and southern hemispheres . BMC Evolutionary Biology , 12 ( 1 ), 244 . doi: 10.1186/1471-2148-12-244 OpenUrl CrossRef PubMed 110. ↵ Wake , D. B . ( 2009 ). What Salamanders Have Taught Us About Evolution . Annual Review of Ecology, Evolution, and Systematics , 40 ( 1 ), 333 – 352 . doi: 10.1146/annurev.ecolsys.39.110707.173552 OpenUrl CrossRef 111. ↵ Wang , I. J . ( 2013 ). Examining the full effects of landscape heterogeneity on spatial genetic variation: a multiple matrix regression approach for quantifying geographic and ecological isolation: special section . Evolution , 67 ( 12 ), 3403 – 3411 . doi: 10.1111/evo.12134 OpenUrl CrossRef PubMed Web of Science 112. ↵ Wang , I. J. , & Bradburd , G. S . ( 2014 ). Isolation by environment . Molecular Ecology , 23 ( 23 ), 5649 – 5662 . doi: 10.1111/mec.12938 OpenUrl CrossRef 113. ↵ Waples , R. S . ( 2022 ). What is N e , anyway? Journal of Heredity , 113 ( 4 ), 371 – 379 . doi: 10.1093/jhered/esac023 OpenUrl CrossRef PubMed 114. ↵ Wickham , H . ( 2016 ). ggplot2: Elegant Graphics for Data Analysis . Springer-Verlag New York . 115. ↵ Wielstra , B. , Crnobrnja-Isailović , J. , Litvinchuk , S. N. , Reijnen , B. T. , Skidmore , A. K. , Sotiropoulos , K. , Toxopeus , A. G. , Tzankov , N. , Vukov , T. , & Arntzen , J. W. ( 2013 ). Tracing glacial refugia of Triturus newts based on mitochondrial DNA phylogeography and species distribution modeling . Frontiers in Zoology , 10 ( 1 ), 13 . doi: 10.1186/1742-9994-10-13 OpenUrl CrossRef PubMed 116. ↵ Wright , S . ( 1931 ). Evolution in Mendelian populations . Genetics , 16 ( 2 ), 97 – 159 . doi: 10.1093/genetics/16.2.97 OpenUrl FREE Full Text 117. ↵ Wright , S . ( 1943 ). Isolation by distance . Genetics , 28 ( 2 ), 114 – 138 . doi: 10.1093/genetics/28.2.114 OpenUrl FREE Full Text 118. ↵ Wright , S. D. , Gillman , L. N. , Ross , H. A. , & Keeling , D. J . ( 2010 ). Energy and the tempo of evolution in amphibians . Global Ecology and Biogeography , 19 ( 5 ), 733 – 740 . doi: 10.1111/j.1466-8238.2010.00549.x OpenUrl CrossRef 119. ↵ Yang , Z . ( 2014 ). Molecular evolution: a statistical approach . Oxford University Press , Oxford . 120. ↵ Zamudio , K. R. & Wieczorek , A. M. ( 2007 ). Fine-scale spatial genetic structure and dispersal among spotted salamander (Ambystoma maculatum) breeding populations . Molecular Ecology , 16 : 257 – 274 . doi: 10.1111/j.1365-294X.2006.03139.x OpenUrl CrossRef PubMed View the discussion thread. Back to top Previous Next Posted November 10, 2025. Download PDF Data/Code Email Thank you for your interest in spreading the word about bioRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Caudata macrogenetics: Geographic attributes and lineage age predict global patterns of mitochondrial genetic variation in salamanders Message Subject (Your Name) has forwarded a page to you from bioRxiv Message Body (Your Name) thought you would like to see this page from the bioRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Caudata macrogenetics: Geographic attributes and lineage age predict global patterns of mitochondrial genetic variation in salamanders Luis Amador , Daniele L.F. Wiley , Irvin Arroyo-Torres , Chris X. McDaniels , Esteban O. Rosario-Sanchez , Hannah Farmer , Hannah Bradley , James Erdmann , Tara A Pelletier , Lisa N Barrow bioRxiv 2025.01.24.634750; doi: https://doi.org/10.1101/2025.01.24.634750 Share This Article: Copy Citation Tools Caudata macrogenetics: Geographic attributes and lineage age predict global patterns of mitochondrial genetic variation in salamanders Luis Amador , Daniele L.F. Wiley , Irvin Arroyo-Torres , Chris X. McDaniels , Esteban O. Rosario-Sanchez , Hannah Farmer , Hannah Bradley , James Erdmann , Tara A Pelletier , Lisa N Barrow bioRxiv 2025.01.24.634750; doi: https://doi.org/10.1101/2025.01.24.634750 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Evolutionary Biology Subject Areas All Articles Animal Behavior and Cognition (7629) Biochemistry (17660) Bioengineering (13881) Bioinformatics (41909) Biophysics (21435) Cancer Biology (18576) Cell Biology (25479) Clinical Trials (138) Developmental Biology (13366) Ecology (19887) Epidemiology (2067) Evolutionary Biology (24301) Genetics (15598) Genomics (22482) Immunology (17726) Microbiology (40359) Molecular Biology (17162) Neuroscience (88529) Paleontology (666) Pathology (2830) Pharmacology and Toxicology (4820) Physiology (7636) Plant Biology (15125) Scientific Communication and Education (2044) Synthetic Biology (4290) Systems Biology (9817) Zoology (2269)