https://doi.org/10.5061/dryad.fj6q5744h

Voris_et_al-TyrannosauroideaCharacterMatrix.xslx: Copy of the phylogenetic character datasheet in Microsoft Excel .xlsx format.

Voris_et_al-TyrannosauroideaCharacterMatrix.csv: Copy of the phylogenetic character datasheet in comma delimited .csv format. Contains the same data as Voris_et_al-TyrannosauroideaCharacterMatrix.xlsx.

Voris_et_al-TyrannosauroideaCharacterMatrix.nex: Copy of the phylogenetic character datasheet in Nexus .nex format. Contains the same data as Voris_et_al-TyrannosauroideaCharacterMatrix.xlsx.

Voris_and Zelenitsky_et_al-Illustrated_Character_List.pdf (Zenodo): Character state illustrations.

Voriset_al_SupportingCode_Scripts.zip (Zenodo): Scripts and additional data needed to replicate our results.

NOTE: All instances of the placeholder “NEWSPECIES” refer to the new Bayanshiree tyrannosauroid: Khankhuuluu mongoliensis.

For additional correspondence or questions email Jared Voris (jared.voris@ucalgary.ca) or Darla Zelenitsky (dkzeleni@ucalgary.ca)

Descriptions

File: Voris_et_al-TyrannosauroideaCharacterMatrix.xslx

Phylogenetic character datasheet with character score values as indicated in the character list of the supplementary information that accompanies the main text. Includes notes and taxon score modifications (changes indicated in red text) from previous phylogenetic analyses. Cells contain notes in some cases. NOTE: A .csv version is provided below for those unable to open this version.

• Column A: Operational Taxonomic Unit (OTUs) formatted as “CLADE” (i.e., ORNITHOMOSAURIA, MEGARAPTORA), *Genusspecies, *or “Genus QUALIFIER SPECIMEN-NUMBER” in cases where individual specimens were codified for extended phylogenetic tests (e.g., Gorgosaurus juvenile TMP 2009.12.14).
• Row 1: Merged cell grouping characters of the same bone for ease of navigation through the dataset (e.g., Premaxilla, Maxilla, Nasal, etc.).
• Row 2: Character number, indicated in the Supplementary Information that accompanies the main text. Individual cells contain notes providing a brief summation of the character description for navigation.

*Discrete character variables coded as 0, 1, 2… for known and scorable conditions (as per the criteria of individual characters listed in the supplementary information) or ? in cases where the character is unknown or unscorable in each OTU.

File: Voris_et_al-TyrannosauroideaCharacterMatrix.csv

Phylogenetic character datasheet with character score values as indicated in the character list of the supplementary information that accompanies the main text. Simplified .csv version. Does not include any merged cells, cell notes, highlighting, or text colors.

• Column A: Operational Taxonomic Unit (OTUs) formatted as “CLADE” (i.e., ORNITHOMOSAURIA, MEGARAPTORA), Genusspecies, or “Genus QUALIFIER SPECIMEN-NUMBER” in cases where individual specimens were codified for extended phylogenetic tests (e.g., Gorgosaurus juvenile TMP 2009.12.14).
• Row 1: Character number, indicated in the Supplementary Information that accompanies the main text.

*Discrete character variables coded as 0, 1, 2… for known and scorable conditions (as per the criteria of individual characters listed in the supplementary information) or ? in cases where the character is unknown or unscorable in each OTU.

File: Voris_et_al-TyrannosauroideaCharacterMatrix.nex

Phylogenetic character datasheet with character score values as indicated in the character list of the supplementary information that accompanies the main text. Nexus format. Does not include any merged cells, cell notes, highlighting, or text colors. Column and row labels as for Voris_et_al-TyrannosauroideaCharacterMatrix.csv.

*Discrete character variables coded as 0, 1, 2… for known and scorable conditions (as per the criteria of individual characters listed in the supplementary information) or ? in cases where the character is unknown or unscorable in each OTU.

File: Voris_and Zelenitsky_et_al-Illustrated_Character_List.pdf (Zenodo)

Variant character list utilized in our phylogenetic analysis with illustrations provided for most individual characters states. The end of the document includes an index of the specimens illustrated in each character and the corresponding references for/authors of illustrations when not using our own images.

File: Supporting_Data_and_Code.zip (Zenodo)

Contains a spreadsheet document providing measurements and methods that rationalize character state cutoff values for absolute measurements and proportional/ratio characters, a log file from all of our Templeton’s tests, and R scripts and supporting data used to replicate our stochastically estimated time calibration results and biogeographic ancestral state reconstructions as well as output files from our final published runs. Further details provided for each file.

File Contents:

Voris et al-RatioCharacterCutoffs.xlsx - Microsoft Excel spreadsheet establishing cutoff values for measurement and proportional characters in our character dataset using Jenk’s Natural Breaks to identify most suitable gaps for discretization of measurement/ratiometric character states. Jenk’s Natural Breaks Optimizations performed in Microsoft Excel using the Real Statistics Resource Pack (accessible at https://real-statistics.com/free-download/real-statistics-resource-pack/). Spreadsheet is subdivided into sections for each relevant module/bone (e.g., Skull, Maxilla, Mandible, etc.) with accompanying goodness of variance fit calculations for each section. Additional formatting as follows:

Columns

• Genus: Genus of measured individual (e.g., Tyrannosaurus)
• Species: Genus (abbreviated) and species of measured individual (e.g., T. rex)
• Specimen: Specimen number of the measured individual (institutional abbreviations provided below)

• Additional columns correspond to measurements of individuals features or ratios between measurements. Bold face numbers followed by descriptor correspond to measurements for the character states themselves (e.g., 2. Maxilla depth/alveolar row length).

  Jenk’s Natural Breaks Optimization goodness of variance fit calculations located to the right of each bone section with bold numbers indicating the corresponding character. Additional notes provided in cells.

  *All absolute measurements provided in millimeters

  **The cutoff for Jenks Breaks Optimizations were followed following a goodness of variance fit (GVF) value of 0.75 unless otherwise specified in individual characters

  ***Contains necessary calculations and cell formatting that were performed in and may only properly open in Microsoft Excel.

Voris et al-Templetons_Test LOG.txt: Log file output from TNT v. 1.6 following performance of various Templeton’s tests to assess significance of various tree topology rearrangements relative to our single recovered most parsimonious tree. Specific rearrangements tested indicated in supplementary information that accompanies main text. Templeton’s Test script used accessible from https://www.researchgate.net/publication/306055781_TNT_script_for_the_Templeton_Test

FOLDER: Voris Time Calibration Code- Input and scripts used to perform our time calibration and ancestral state estimation analyses as well as output files from these analyses

CONTENTS

• Voris TIMESCALING 3Cal (modGroh et al 2022)_1Ma Bins_2StepSubopt.R - R script used to estimate divergence times using three-rate-calibrated (Cal3) time-scaling method.
• TIMESCALING & ANCESTRAL STATE CODE.R - R script used to scale the phylogenetic tree against 3cal recovered age dates.
• .Rhistory - R history file from published run

• Node_ages.csv - Median node age calculated FOLLOWING cal3 time scaling analysis that were used for time scaling nodes in our phylogenetic tree graphic. Input file for “TIMESCALING & ANCESTRAL STATE CODE.R”

  Columns

  • A. Node number - can be visualized in R using the “nodelabels()” command when a tree is loaded into the R environment. Also indicated in “scaled_tree.pdf” contained in same folder.
  • B. Node age - median node age calculated from cal3 analysis for the corresponding node number

• TIP_ages.csv - Taxa (tip numbers) and first and last appearance datum ages.

  Columns

  • A. OTU - Formatted as Genusspecies (e.g., Tyrannosaurusrex)
  • B. FAD - First Appearance Datum of that species in the fossil record (number of millions of years [Ma])
  • C. LAD - Last Appearance Datum of that species in the fossil record (Ma)

• TIP_agesMIN.csv - Taxa (tip numbers) and first appearance datum age only. INPUT file for “TIMESCALING & ANCESTRAL STATE CODE.R”

  Columns

  • A. Tip number - OTU tip label from trees, can be visualized using the “tiplabels()” command when a tree is loaded in the environment
  • B. FAD of each tree tip (Ma)
• Tyranno125trees_Sub2.phy - Newick tree file containing 125 trees of up to 2 steps suboptimal (used to incorporate phylogenetic uncertainty into node age and ancestral state state estimations). INPUT file for “Voris TIMESCALING 3Cal (modGroh et al 2022)_1Ma Bins_2StepSubopt.R”
• TyrannoMPTree.phy - Single most parsimonious tree from TNT analysis - INPUT file for “TIMESCALING & ANCESTRAL STATE CODE.R”
• TyrannosaurData.RData - RData file containing taxon time bins used in 3cal analysis.

• TyrannoPhyloData.csv - OTU temporal (first and last appearance datum) and biogeographic data as well as references for age dates.

  Columns

  • A. Species - Formatted as Genusspecies (e.g., Tyrannosaurusrex)
  • B. FAD - First Appearance Datum of that species in the fossil record (Ma)
  • C. LAD - Last Appearance Datum of that species in the fossil record (Ma)
  • D. Time Period - Geologic Time Scale age range of that OTU
  • Singleton - categorical “yes” or “no” of whether that OTU is treated as a singleton (NOTE: we use singleton to also define two or more individuals from the same stratigraphic bed)
  • Analysis Time Bin Broad - Categorical qualifier for the age range of the OTU for use in the cal3 analysis
  • first_int - Numerical qualifier for the first instance of the OTU; used in the cal3 analysis
  • last_int - Numerical qualifier for the last instance of the OTU; used in the cal3 analysis - Note: last_int is not the same as last appearance datum. Last_int is different from first_int only in situations where the OTU age is not well-constrained.
  • Biogeography - Categorical qualifier for OTU Biogeographic range
  • Analysis Biogeography Code - Numerical qualifier for OTU Biogeographic Range (1=North America; 2=Europe; 3=Asia)
  • Reference for Stratigraphic Age - Literature reference for OTU stratigraphic ages
    *Literature Cited provided below OTU data rows
• TyrannoGeoscaleTree.pdf - Phylogenetic tree timescaled using 3cal and overlain on the 2012 version of the geologic timescale using the strap R package. OUTPUT from “TIMESCALING & ANCESTRAL STATE CODE.R”. NOTE: The age dates used by the Strap R package to define time bins within the geologic time scale graphic are based on the out-of-date 2012 version of the Geologic Time Scale. Our analyses use time bin age dates from the newer 2020 version of the Geologic Time Scale. This is, however, only a graphical issue and does not affect the results. To replicate our results please compare to time bins from the 2020 Geologic Time Scale.
• FOLDER: OutputCal3-Moros - OUPUT files from our own runs of the stochastic tip calibration and ancestral state mapping. Contains a range of resultant age data files and scaled trees
  • Ages.csv - Analysis estimated node age dates
  • AgesSD.csv - standard deviation of OTU tip ages
  • AgesSummary.csv - Summary of node ages including minimum, qst quartile, median, mean, 3rd quartile, and maximum
  • AllNodeAges.csv - Output for all estimated node ages .csv format
  • AllNodeAges.txt - Output for all estimated node ages .txt format
  • AnalysisAgesCIandMean.csv - Output for summary stats of node ages including rows corresponding to mean, median, and left and right 95% confidence tails
  • AnalysisAgesHistograms.pdf - Output pdf document of distributions of ages for each node

  • Ancestral_biogeo.csv - Likelihood estimations of the ancestral biogeography for each node and tip

    Columns

    • A. Node number and taxon tip labels - Node numbers can be visualized in R using the “nodelabels()” command when a tree is loaded into the R environment. Also indicated in “scaled_tree.pdf” contained in same folder
    • B. 1 - Likelihood North American (x/1)
    • C. 2 - Likelihood European (x/1)
    • D. 3- Likelihood Asian - (x/1)
  • scaled_tree.pdf - Phylogenetic tree timescaled using the sample of cal3 trees. Contains node number labels as referred to in various files in this folder. pdf format. OUTPUT from “TIMESCALING & ANCESTRAL STATE CODE.R”
  • Scaled_tree.tre - Phylogenetic tree timescaled using the sample of cal3 trees. .tre format. OUTPUT from “TIMESCALING & ANCESTRAL STATE CODE.R”
  • TyrannoBioGeoSIMMAP-BROAD.pdf - Timescaled phylogenetic tree with piecharts at nodes indicating likelihood biogeographic ancestry. OUTPUT from “TIMESCALING & ANCESTRAL STATE CODE.R”
• FOLDER: Bayes - Input and output files for ancestral state mapping of our fossilized birth-death (FBD) time calibrated trees. Contains three file folders corresponding to three different MPT topologies. Contents of folders are the same and vary only in starting tree topology. ## in file name will differ between runs but follows same format (i.e., 1, 2, 3)
  • bayes##.tre - Unscaled tree topology
  • BAYES##_TIMESCALING & ANCESTRAL STATE CODE.R - R script used to scale the phylogenetic tree against FBD age dates extracted from MrBayes output tree files

  • BioGeoBayes##.csv - Spreadsheet containing tip states for biogeography ancestral state reconstruction

    Columns

    • A. Species - Formatted as Genusspecies (e.g., Tyrannosaurusrex)
    • B. Species biogeography - 1=North America, 2=Europe, 3=Asia, 4=Arbitrary bin for extant taxon anchor biogeography i.e., “Gallus”
  • MrBayesFBDTree##_TyrannoAllCompCon.tre - Output timescaled tree file from MrBayes FBD (used to extract ages)

  • Node_agesBAYES##.csv - Node calculated FOLLOWING FBD time scaling analysis. Input file for “BAYES##_TIMESCALING & ANCESTRAL STATE CODE.R”

    Columns

    • A. Node number - can be visualized in R using the “nodelabels()” command when a tree is loaded into the R environment.
    • B. Node age - median node age calculated from cal3 analysis for the corresponding node number

  • TIP_agesBAYES##.csv - TIP_ages.csv - OTU (tip numbers) and first appearance datum ages.

    Columns

    • A. OTU Tip Number - can be visualized in R using the “tiplabels()” command when the tree is loaded into the R environment
    • B. FAD - First Appearance Datum of that species in the fossil record (number of millions of years [Ma])
  • Tyrannosauroidea FBD Normal Parismony-##.nex - MrBayes input Nexus file containing OTU score data, tree constraints (as trees were constrained to MPT topologies from parsimony analyses), and OTU age range
  • FOLDER: Output - Output data timescaling and ancestral state files from “BAYES##_TIMESCALING & ANCESTRAL STATE CODE.R”

    • Ancestral_biogeo##.csv - Ancestral_biogeo.csv - Likelihood estimations of the ancestral biogeography for each node

      Columns

      • A. Node number and taxon tip labels - Node numbers can be visualized in R using the “nodelabels()” command when a tree is loaded into the R environment. Also indicated in “scaled_tree.pdf” contained in same folder
      • B. 1 - Likelihood North American (x/1)
      • C. 2 - Likelihood European (x/1)
      • D. 3- Likelihood Asian - (x/1)
    • BAYES##scaled_tree.pdf - Tree scaled to age dates produced by FBD analysis. pdf format
    • BAYES##scaled_tree.pdf - Tree scaled to age dates produced by FBD analysis. tre format
    • BAYES##timescaled_tree.pdf - Tree scaled to age dates produced by FBD analysis overlain on 2012 Geologic Timescale

Institutional Abbreviations. AMNH, American Museum of Natural History, New York, NY, U.S.A.; BDM, Badlands Dinosaur Museum, Dickinson, SD, U.S.A.; БTM, Natural History Museum of Mongolia, Ulaanbaatar, Mongolia; BYU, Brigham Young University Museum of Paleontology, Provo, UT, U.S.A.; CMN, Canadian Museum of Nature, Ottawa, ON, Canada; CMNH, Cleveland Museum of Natural History, Cleveland, OH, U.S.A.; CPC, Coleccion Paleontologica de Coahuila, Museo del Desierto, Saltillo, Coahuila, Mexico; FMNH, Field Museum of Natural History, Chicago, IL, U.S.A.;FRDC, Fossil Research and Development Center, Gansu Bureau of Geology and Mineral Resources Exploration, Lanzhou, P. R. China; GM, Ganzhou Museum, Ganzhou, China; IVPP, Institute of Vertebrate Paleontology and Paleoanthropology, Beijing, China;MOR, Museum of the Rockies, Bozeman, MT, U.S.A.; MPC, Institute of Paleontology, Mongolian Academy of Sciences, Ulaanbaatar, Mongolia; NMMNH, New Mexico Museum of Natural History, Albuquerque, NM, U.S.A.; PIN, Paleontological Institute, Moscow, Russia; RMM, Red Mountain Museum (curated at the McWane Science Center), Birmingham, AL, U. S. A.; RSM, Royal Saskatchewan Museum, Regina, SK, Canada; SMG, Shanxi Museum of Geology, Taiyuan, Shanxi Province, China; TMP, Royal Tyrrell Museum of Palaeontology, Drumheller, AB, Canada; UALVP, University of Alberta Laboratory of Vertebrate Paleontology, Edmonton, AB, Canada; UM NH, Utah Museum of Nautral History, Salt Lake City, UT, U.S.A. ZIN PH, Paleoherpetological Collection,  Zoological Institute, Russian Academy of Sciences, Saint Petersburg, Russia.

NOTES: All instances of the placeholder “NEWSPECIES” refer to the newly Bayanshiree tyrannosauroid: Khankhuuluu mongoliensis.

Code/software

All analyses were performed in R version 4.3.3 with the following installed packages: paleotree 3.4.7, strap 1.6-1, phytools 2.3-0, foreach 1.5.2, doParallel 1.0.17, geiger 2.0.11, matrixStats 1.4.1, dplyr 1.1.4, TreeTools 1.12.0.