Data from: Evolution and biogeography of the widespread dragonfly genus Orthetrum (Odonata: Anisoptera: Libellulidae): Massive dispersal in the tertiary
Data files
Jul 01, 2026 version files 800.59 MB
-
README.md
16.40 KB
-
Supplemental_Material_.zip
800.57 MB
Abstract
With 67 species, Orthetrum Newman, 1833 is the most speciose genus within the family Libellulidae Leach, 1815 (Odonata: Anisoptera), distributed across the Afrotropical, Palearctic, Indo-Malayan, Malagasy, Oceania, Wallacea, and Australasian regions. Despite its wide distribution, serving as a top predator in freshwater ecosystems, the evolutionary history of Orthetrum has remained unresolved, with prior studies limited to morphology and sparse genetic sampling (<33% of species). We present here the most comprehensive phylogenetic study of Orthetrum to date, sampling 50 of 67 species (~75%) using Anchored Hybrid Enrichment (AHE). Our analyses, integrating concatenated maximum likelihood and coalescent-based approaches, provide a robust, strongly supported phylogeny that confirms the monophyly of Orthetrum and reveals four distinct clades. Two clades comprise mostly Afrotropical species, whereas the remaining clades are from the Palearctic, Indo-Malayan, Oceania, Wallacea, and Australasian regions. Divergence time estimates indicate that Orthetrum originated ~32 million years ago, with major diversification occurring from 28 Ma onward, well after major continental fragmentation ~events. Ancestral range reconstruction under the best-supported dispersal–extinction–cladogenesis (DEC) framework does not support a single-region origin; instead, early lineages are inferred to have occupied a broadly connected Afro–Eurasian system. Subsequent diversification is characterized by stepwise dispersal across adjacent regions, including repeated expansions into the Indo-Malayan and Palearctic regions, followed by colonization of Australasia, Wallacea, and Oceania. The Afrotropical region emerges as a major center of lineage persistence and diversification, with subsequent range expansion in some lineages during the early Miocene (~23 Ma). Madagascar reflects at least five independent colonization events from Africa, resulting in one species endemic to the island.
We have submitted:
A compressed file named: Supplemental_Material_.zip
Description:
This Supplemental material.zip contains the datasets and analysis files supporting the manuscript on the phylogenomics, divergence time estimation, and historical biogeography of Orthetrum (Odonata: Libellulidae). The data include sequence alignments, phylogenetic trees, BioGeoBEARS input and output files, divergence time estimation files, and supplementary datasets used in the study.
Folder contents:
- Alignments/
========================================
Contains Assemblied_Raw_data_ AHE reads of Orthetrum species + outgroups, concatenated_alignment.fas,partition_def.txt from our AHE pipeline. We followed methods outlined in Breinholt et al. (2018) to assemble and assign orthology to each target capture locus (Supplemental_Table_S3.csv)
concatenated_alignment.fas*: Concatenated nucleotide alignment matrix used for phylogenetic inference analyses
partition_def.txt*: Partition definition file specifying locus boundaries and partitioning scheme used in IQ-TREE analyses
- assemblies_raw_data/
Folder containing raw assembled sequence data generated during the Anchored Hybrid Enrichment (AHE) pipeline
- Phylogenetic analysis/
========================================
Contains phylogenetic datasets and tree files generated during sequence alignment, concatenation, and phylogenetic inference analyses.
- Astral_tree.tre**: Species tree estimated using ASTRAL-III
- concord.cf.tree**: Concordance factor phylogenetic tree generated from IQ-TREE analyses
- concord.cf.tree.nex**: Nexus-format concordance factor tree file
concord.cf_the_full_quartet_frequencies.stat*: Statistical output file containing quartet concordance frequencies and support values generated during concordance factor analyses in IQ-TREE
- FcC_smatrix.fas**: Concatenated nucleotide alignment matrix used for phylogenetic analyses
- odo_concat.contree**: Consensus phylogenetic tree generated from concatenated analyses
- Divergence_time_estimation_using_a_secondary_calibration/
========================================
This directory contains all files associated with divergence-time estimation analyses conducted in MCMCTree under alternative secondary calibration schemes derived from Newton et al. (2026). The analyses were performed to evaluate the robustness of divergence-time estimates to different calibration strategies.
- The directory is organized into two subdirectories:
mcmc_with_Orthetrum_calibration/– analyses incorporating an Orthetrum crown-age calibration in addition to deeper libellulid secondary calibrations.mcmc_no_Orthetrum_calibration/– analyses using only deeper libellulid secondary calibrations and excluding the Orthetrum crown-age calibration.
Within each calibration scheme, four independent MCMCtree analyses (RUN_1–RUN_4) were conducted using different random seeds to assess convergence and consistency among runs. Exploratory analyses (RUN_1 and RUN_2) were used to evaluate MCMC performance and parameter settings, whereas final production analyses (RUN_3 and RUN_4) were used for posterior inference and divergence-time estimation.
Each run directory contains the input files, control files, approximate-likelihood files, posterior samples, log files, and summarized chronograms generated during the MCMCtree analyses, including:
FcC_smatrix.fas– concatenated sequence alignment used for divergence-time estimation.Orthetrum_Ultrametric.newick– calibrated guide tree with node-age constraints.Orthetrum_noCal.newick-Calibrated guide tree with no Orthetrum node_age constraints.mcmctree_Modified.ctl– MCMCtree control file specifying priors and MCMC settings formcmc_with_Orthetrum_calibration/mcmctree_noCal.ctl-MCMCtree control file specifying priors and MCMC settings formcmc_no_Orthetrum_calibration/in.BVandout.BV– approximate-likelihood input and output files.mcmc.txt– posterior samples of divergence times and model parameters.out– main MCMCtree output file containing posterior summaries.run*.log– analysis log files.SeedUsed– random seed used for each run.FigTree.tree– consensus chronogram formatted for visualization in FigTree.
Posterior estimates reported in the manuscript were based on the final production runs after confirming convergence among independent analyses.
- BioGeoBEARS_Biogeography_analyses/
========================================
Contains input files, scripts, model outputs, ancestral range reconstructions, and model comparison results generated from BioGeoBEARS analyses under both unconstrained and constrained biogeographic frameworks.
Key files include:
- allowed_states_constrained.txt
List of geographically constrained ancestral range combinations permitted during constrained BioGeoBEARS analyses - RUN4_baseline_BAYAREALIKE_results.RData
Baseline unconstrained BAYAREALIKE model output - RUN4_baseline_DIVALIKE_results.RData
Baseline unconstrained DIVALIKE model output - RUN4_final_baseline_DEC_results.RData
Baseline unconstrained DEC model output - RUN4_baseline_final_analysis_with_root_support.R
R script used for unconstrained ancestral range reconstruction and root state evaluation - biogeography_pipeline_constrained.R
R script implementing constrained BioGeoBEARS analyses and model comparisons - RUN4_DEC_constrained_sabina_ABC_results.RData
DEC model output under constrained coding of Orthetrum sabina using the ABC framework - RUN4_DEC_constrained_sabina_DEF_results.RData
DEC model output under constrained coding of Orthetrum sabina using the DEF framework - RUN4_DIVALIKE_constrained_sabina_ABC_results.RData
DIVALIKE model output under constrained ABC coding - RUN4_DIVALIKE_constrained_sabina_DEF_results.RData
DIVALIKE model output under constrained DEF coding - RUN4_BAYAREALIKE_constrained_sabina_ABC_results.RData
BAYAREALIKE model output under constrained ABC coding - RUN4_BAYAREALIKE_constrained_sabina_DEF_results.RData
BAYAREALIKE model output under constrained DEF coding - constrained_sabina_ABC_model_comparison.csv
Model comparison table for constrained ABC analyses - constrained_sabina_DEF_model_comparison.csv
Model comparison table for constrained DEF analyses - Fig_RUN4_constrained_BAYAREALIKE_sabina_DEF.pdf
Ancestral range reconstruction figure for constrained BAYAREALIKE DEF analysis - Fig_RUN4_constrained_DIVALIKE_sabina_DEF.pdf
Ancestral range reconstruction figure for constrained DIVALIKE DEF analysis - Fig_RUN4_DEC_constrained_sabina_DEF.pdf
Ancestral range reconstruction figure for constrained DEC DEF analysis - Fig_RUN4_DIVALIKE_pie_tree.pdf
Pie-chart ancestral range reconstruction under unconstrained DIVALIKE model - Fig_RUN4_BAYAREALIKE_pie_tree.pdf
Pie-chart ancestral range reconstruction under unconstrained BAYAREALIKE model - Fig_RUN4_model_comparison_AICc.pdf
Model comparison figure summarizing AICc support across models - Fig_RUN4_constrained_sabina_DEF_model_comparison_AICc.pdf
Model comparison figure for constrained DEF analyses - final_DEC_root_focus_A_AB_ABC.csv
Summary of focal root-state probabilities under DEC analyses - final_DEC_root_top_states.csv
Summary of highest-probability ancestral states inferred under DEC analyses - Geographical_areas_matched_sabina_ABC.data / .txt
Geographic coding files for constrained ABC representation of Orthetrum sabina - Geographical_areas_matched_sabina_DEF.data / .txt
Geographic coding files for constrained DEF representation of Orthetrum sabina - Orthetrum_MCMCtree_RUN4.tre
Time-calibrated phylogenetic tree used for BioGeoBEARS analyses - Orthetrum_no_outgroups.tre
Input phylogeny excluding outgroup taxa used for ancestral range estimation - Orthetrum_DEC_ancestral_ranges.pdf
Final DEC ancestral range reconstruction figure presented in the manuscript
Biogeographic regions:
A = Afrotropical
B = Indo-Malayan
C = Palearctic
D = Australasia
E = Oceania
F = Wallacea
G = Malagasy
- Supplementary_tables/
========================================
Contains supplementary tables referenced in the manuscript
- Table S1: Taxon assigned sampling regions for BioGeoBEARS.
- Table S2: Species absent from this study, with and their likely position within the phylogenetic tree based on morphology and biogeographical region.
- Table S3: Probe set, AHE Loci recovered, Gap_ ambiguity & mean contig length.
- Table S4. Gene concordance factor (gCF), site concordance factor (sCF), and quartet score statistics for all internal nodes in the Orthetrum phylogeny inferred from the concatenated maximum-likelihood analysis. Values are presented for each node together with the corresponding quartet frequencies and concordance categories, where values <30% indicate low concordance, 30–60% indicate moderate concordance, and >60% indicate high concordance.
- Table S5. BioGeoBEARS likelihoods under different biogeographic models with unconstrained analysis.
- Table S6. BioGeoBEARS likelihoods under different biogeographic models with constrained_ O. sabina_ABC (Afro-Eurasian regions) _model comparison.
- Table S7. BioGeoBEARS likelihoods under different biogeographic models under constrained_ O. sabina_DEF (Indo-Australasian regions) _model_comparison.
- Table S8. Ancestral range probabilities under the unconstrained DEC model.
- Table S9. Ancestral range probabilities under the DEF-constrained DEC model.
- Table S10. Ancestral range probabilities under the ABC-constrained DEC model.
- Table S11.Divergence time estimates for all nodes (59–115) in the Orthetrum chronogram, including node identity, crown or stem designation, mean age estimates (Ma), 95% highest posterior density (HPD) intervals, and corresponding geological periods inferred from the MCMCtree analysis
- Figures/
========================================
Contains figure referenced in the manuscript
Figure S1* Supplementary maximum-likelihood (ML) with branch length
Recommended Software for Data Analysis
Reproducing the analyses included in this repository requires the following software packages:
- Analyses contained in 1_Alignments/ and 3_Phylogenetic_analysis/ were conducted using IQ-TREE2: (http://www.iqtree.org/)
- Divergence time estimation analyses contained in 4_Divergence_time_estimation_using_a_secondary_calibration/ were performed using the PAML package, specifically the MCMCtree program: (http://abacus.gene.ucl.ac.uk/software/paml.html)
- Biogeographic analyses contained in 5_BioGeoBEARS_Biogeography_analyses/ were conducted in R and RStudio using the BioGeoBEARS package: R / RStudio: https://posit.co/download/rstudio-desktop/ BioGeoBEARS: https://github.com/nmatzke/BioGeoBEARS
Data Formats
Spreadsheet Files:
- .csv: Comma-separated value files containing model comparison statistics, ancestral state summaries, and supplementary analytical outputs from BioGeoBEARS analyses (e.g., constrained_sabina_ABC_model_comparison.csv, constrained_sabina_DEF_model_comparison.csv, final_DEC_root_top_states.csv).
Text Files:
- .txt: Plain text files containing partition definitions, geographic range coding schemes, constrained ancestral state combinations, MCMCtree outputs, and additional analytical metadata used in phylogenetic, divergence time, and biogeographic analyses (e.g., partition_def.txt, allowed_states_constrained.txt, Geographical_areas_matched_sabina_ABC.txt, Geographical_areas_matched_sabina_DEF.txt, mcmc.txt, and tmp0001.txt).
R Scripts:
- .R: R scripts used to perform ancestral range reconstruction analyses and model comparisons in BioGeoBEARS (e.g., baseline_final_analysis_with_root_support.R and biogeography_pipeline_constrained.R).
Cluster Job Files:
- .job: SLURM job scripts submitted to the high-performance computing cluster, containing commands and parameters used for divergence time estimation analyses (e.g., mcmctree.job).
Phylogenetic Files:
- .tre: Phylogenetic tree files generated from IQ-TREE, ASTRAL, and MCMCtree analyses, viewable using FigTree (http://tree.bio.ed.ac.uk/software/figtree/) or similar tree visualization software (e.g., Astral_tree.tre, odo_concat.contree, MCMCtree_tree.tre, and Orthetrum_no_outgroups.tre).
- .newick: Newick-format phylogenetic trees used as input for divergence time estimation analyses and ancestral range reconstruction (e.g., Orthetrum_noCal.newick and Orthetrum_Ultrametric.newick).
Alignment Files:
- .fas / .fasta: FASTA-format nucleotide alignment files containing concatenated Anchored Hybrid Enrichment (AHE) sequence datasets used in phylogenetic and divergence time analyses. These files can be opened using AliView (https://ormbunkar.se/aliview/) or Mesquite (https://www.mesquiteproject.org/) (e.g., concatenated_alignment.fas and FcC_smatrix.fas).
Time-Divergence Files:
- .ctl: Control files defining model settings and parameters for divergence time estimation analyses performed using MCMCtree in the PAML package (e.g., mcmctree_noCal.ctl and mcmctree_Modified.ctl).
- .BV: Approximate likelihood files containing branch length variance information generated during MCMCtree analyses (e.g., in.BV and out.BV).
RData Files:
- .RData: Serialized R workspace files containing BioGeoBEARS model outputs and ancestral range reconstruction results (e.g., final_baseline_DEC_results.RData, DEC_constrained_sabina_DEF_results.RData, and baseline_DIVALIKE_results.RData).
Additional Files:
- .pdf: Portable Document Format files containing ancestral range reconstruction figures, model comparison plots, and supplementary analytical figures generated during BioGeoBEARS analyses (e.g., Orthetrum_DEC_ancestral_ranges.pdf and Fig_S3_model_comparison_AICc.pdf).
Notes:
- Time-calibrated phylogenetic analyses were conducted using secondary calibrations derived from Newton et al. (2026).
- Constrained and unconstrained BioGeoBEARS analyses are included.
- The widespread taxon Orthetrum sabina was evaluated under alternative constrained codings as part of sensitivity analyses.
Usage, Compatibility, and Accessibility
========================================
This dataset encompasses a broad range of file formats associated with phylogenetic, divergence time estimation, and historical biogeographic analyses conducted in this study of Orthetrum.
Most files can be viewed or edited using standard text-editing software such as BBEdit (https://www.barebones.com/products/bbedit/index.html), Visual Studio Code, or similar programs. Phylogenetic tree files can be visualized using FigTree (http://tree.bio.ed.ac.uk/software/figtree/), while sequence alignment files can be opened using AliView (https://ormbunkar.se/aliview/) or Mesquite (https://www.mesquiteproject.org/). PDF figures and graphics can be viewed using standard PDF readers or edited using vector graphic software such as Inkscape (https://inkscape.org/).
All scripts, analytical outputs, and associated datasets are publicly available and freely accessible. Researchers are encouraged to use the original file formats to reproduce analyses, evaluate methodological consistency, and facilitate future comparative phylogenomic and biogeographic studies.
Questions regarding dataset organization, analytical workflows, or file interpretation may be directed to the corresponding author. We are committed to supporting reproducibility, transparency, and future research applications of these phylogenomic datasets.
