Data from: "Lower Ordovician synziphosurine reveals early euchelicerate diversity and evolution"

PRESENTATION This README file accompanies the data produced in the course of the prepara9on of the manuscript “Lower Ordovician synziphosurine reveals early euchelicerate diversity and evolu9on” by Lorenzo Lustri, Pierre Gueriau and Allison C. Daley. Data are presented in two separate folders: - The “data_used_in_the_manuscript” folder, which gathers only the data used in the manuscript - The “more_data” folder, which gathers data collected in the course of the project but were not used during the prepara9on of the associated manuscript. This document includes: - Details on the methods used to generate the data - Details on the metadata associated to the data - Usage notes on how to open, run and/or process the provided data METHODS Op/cal photography. The specimens were photographed with an SLR camera (Canon EOS 800D equipped with CANON macro lens MP-E 65 mm 1:2.8 1–5×) mounted on a stand and connected to a focus stacking (z-stacking) system (STACKSHOT 3X), using different combina9on of lightning condi9ons: normal light, polarized light, dry, covered in alcohol. Focus (Z-)stacks were rendered using the Helicon Focus sodware. Synchrotron X-ray computed microtomography. Two specimens (MGL.102637a and MGL.102841) were imaged using synchrotron X-ray microtomography at the X02DA TOMCAT beamline of the Swiss Light Source, Paul Scherrer Ins9tut, Villigen, Switzerland. Measurements were performed using monochroma9c beam of 35 and 18 keV respec9vely, a single propaga9on distance of 250 mm, a 100 µm LuAg:Ce scin9llator, and a 4× objec9ve, yielding reconstructed tomographic data with a voxel size of 1.75 µm. 1501 projec9ons were recorded over 180° with exposure of 400 and 1000 ms respec9vely. Reconstruc9on was performed on a 60-core Linux PC farm using a Fourier transform rou9ne and a regridding procedure, with a Paganin phase retrieval algorithm. The obtained tomograms are provided in this repository as individual .TIFF 8-bit images. Tomograms presented in Figure 4 of the manuscript were processed using ImageJ, manually adjus9ng the mininum and maximum values in order to provide the strongest possible enhancement of contrast. Manual segmenta9on and three-dimensional rendering were performed using the sodware MIMICS Innova9on Suite 19.0 (Materialise) at the IPANEMA laboratory (Gif-sur-Yvepe, France). Please note that the segmenta9on file and exported three-dimensional rendering images are not provided in this repository. This is because segmenta9on (i) was performed using commercial sodware, resul9ng in a file that cannot be opened by any open-source program, and (ii) may vary slightly between different prac99oners, making it prac9cally impossible to reproduce the exact same processing. Therefore, for these files, we invite you to refer to the figures in the associated manuscript. Phylogene/c analyses. The first Bayesian phylogene9c analyses were performed following methods in Aria & Caron (2019, Nature 573, 586–589) using mrBayes ver. 3.2.7a through the Cipres science gateway web site. The analyses consist of tree searches following an Mkv + Γ model with four chains sampling during four runs for 10,000,000 Markov chain Monte Carlo genera9ons, a tree sampled every 1,000 genera9ons and burn-in of 20%. Analyses were constrained with par9al backbone. Two versions of this matrix have been analyzed to test two different characters codings for Mollisonia plenovenatrix and they are available alongside coding for Setapedites abundan4s in this repository. A constraint for the Arachnopulmonata has also been used and the character coding is available in this repository. The second set of Bayesian phylogene9c analyses were performed on the data matrix from Lamsdell (2013, Zoological Journal of the Linnean Society 167, 1–27). We followed the same methods of the previous Bayesian analyses but without par9al backbone constraint. Character and taxa coding and modifica9ons to this matrix are available in this repository. Two versions of this matrix have been analyzed to test different characters codings for Mollisonia plenovenatrix and they are available alongside coding for Setapedites abundan4s in this repository. The first parsimony analyses were performed on the same data matrix modified from Lamsdell (2013, Zoological Journal of the Linnean Society 167, 1–27) and following its methods. The analyses consist of random addi9on sequences followed by branch swapping (100,000 repe99ons), all characters unordered and of equal weight followed by jacknife (33% dele9on, 1000 repe99ons) and bootstrap (50% dele9on, 1000 repe99ons) using TNT ver. 1.5. Character coding for Setapedites abundan4s is available in the electronic supplementary material dataset associated to the manuscript. The second set of parsimony analyses were performed on the same data matrix modified from Lamsdell (2013, Zoological Journal of the Linnean Society 167, 1–27) and following its methods except for an implied weight of 12K using TNT ver. 1.5. A last set of phylogene9c analysis excluding the ar9podans from the data matrix (Olenoides serratus, Emeraldella brocki, and Sidneyia inexpectans) has also been performed and the character coding is available in this repository. Chronological scaling. Chronological data for the 39 taxa included in the Bayesian analyses were collected from PBDB (paleobiodb.org) and from data available in Lamsdell (2013, Zoological Journal of the Linnean Society 167, 1–27). The tree resul9ng from phylogene9c parsimony analyses was ploped against the geological 9mescale using the strap R package in RStudio to obtain the chronograms. The tree was rescaled with command “mbl”. METADATA Op/cal photography. The specimen number and lightning condi9ons are indicated in the file name as follows: “SpecimenNumber_Dry/CoveredinAlcohol_NormalLight/ UnderPolarizedLight&Filter”. For focus (Z-)stacked photographs, “_Zstack_composite” is added at the end of the file name. Scaling is either embedded within the photograph or provided as a separate file, for which the file name indicates the specimen number, the fact that this is a scale image and details about how to read gradua9on, as follows: “SpecimenNumber_scale_Gradua9onValue”. Synchrotron X-ray computed microtomography. The specimen number, instrument used, resolu9on (pixel size), image format and serial number are indicated in the tomogram file name as follows: “SpecimenNumber_X02DA-TOMCAT-SLS_1p75um_8bits_SerialNumber”. The series of tomograms are gathered in a folder with a similar name, in the format of “SpecimenNumber_X02DA-TOMCAT-SLS_1p75um_8bits_SerialNumber”. This folder is associated with an addi9onal README .txt file ("SpecimenNumber_X02DA-TOMCATSLS_1p75um_8bits_README.txt”) that details all parameters used in the collec9on (energy, propaga9on distance, scin9llator, objec9ve, pixel size, number of projec9ons, angular domain, exposure 9me), genera9on and prepara9on of the dataset (reconstruc9on sodware, correc9ons, data reduc9on). Phylogene/c analyses. The type of analysis, Bayesian or parsimony, is indicated in the file name. Chronological scaling. The file to reproduce the chronological scaling is file Setapedites_age.txt. This file includes First and Last Appearance Dates (FAD and LAD, respec9vely) for all the considered taxa. USAGE NOTES Op/cal photography. Simple and focus (Z-)stacked photographs do not require any par9cular method to be open and visualized, they can simply be open and read classically by any computer. Synchrotron X-ray computed microtomography. Individual tomograms, in their .TIFF 8-bit format, do not require any par9cular method to be open and visualized, they can simply be open and read classically by any computer. The series of tomograms can be visualized and possibly reoriented and/or reduced in the freeware ImageJ/FIJI. 3D rendering of the fossil needs to be performed using specific sodware called segmenta9on sodware; in the course of the project segmenta9on and 3D rendering were performed using the commercial sodware MIMICS Innova9on Suite 19.0 (Materialise) at the IPANEMA laboratory (Gif-sur-Yvepe, France), but freeware such as 3D Slicer (hpps://www.slicer.org/) or Dragonfly (hpps://www.theobjects.com/dragonfly) can also be used. Phylogene/c analyses. The Bayesian analyses can be reproduced by running the file Bayesyan_matrix_and_script_SDX.nex in MrBayes, by simply typing “execute Bayesyan_matrix_and_script_SDX.nex” on the MrBayes command prompt. (Note that the SDX in the file name refers to Supplementary Data X (1–5, 7, 8) of the associated manuscript. Equal weight parsimony analyses with their Bremer, Jacknife and Bootstrap support can be reproduced by simply typing the TNT script below into the TNT command prompt to run the analyses on Parsimony_matrix_SD6.tnt or Parsimony_matrix_SD9.tnt (Note that the script below runs file Parsimony_matrix_SD6.tnt; replace SD6 with SD9 throughout the script to run Parsimony_matrix_SD9.tnt): Equal weight analyses and Bremer support mxram 10000 proc Parsimony_matrix_SD6.tnt hold 10000000; outgroup 2; mult: replic 100000 hold 100; subopt 10; mult; taxname= ; export* Parsimony_matrix_SD6_EWAB; BSUPPORT=0.01; TSAVE *bremer; SAVE *bremer; SAVE /; Jacknife and Bootstrap supports cd /Users/ ##Set here the directory where the files have been saved## ./tnt.command mxram 10000 proc Parsimony_matrix_SD6.tnt hold 10000000; outgroup 2; mult: replic 100000 hold 100; mult; nelsen*; taxname= ; export* Parsimony_matrix_SD6_EWJB; ##The name of the exported file here can be changed## RESAMPLE:jak replica9ons 1000[ mult = tbr replic 100 hold 10 ] from nelsen_tree_number; ##Replace nelsen_tree_number with the number of the nelsen tree## RESAMPLE:boot replica9ons 1000[ mult = tbr replic 100 hold 10 ] from nelsen_tree_number; ##Replace nelsen_tree_number with the number of the nelsen tree## Implied weight parsimony analyses with their Bremer, Jacknife and Bootstrap support can be reproduced by simply typing the TNT script below into the TNT command prompt to run the analyses on Parsimony_matrix_SD6.tnt or Parsimony_matrix_SD9.tnt (Note that the script below runs file Parsimony_matrix_SD6.tnt; replace SD6 with SD9 throughout the script to run Parsimony_matrix_SD9.tnt): 12 k implied weight analyses and Bremer support mxram 10000 proc Parsimony_matrix_SD6.tnt hold 10000000; outgroup 2; mult: replic 100000 hold 100; PIWE=12; subopt 10; mult; taxname= ; export* Parsimony_matrix_SD6_12KB; BSUPPORT=1[; TSAVE *bremer; SAVE *bremer; SAVE /; Jacknife and Bootstrap supports cd /Users/ ##Set here the directory where the files have been saved## ./tnt.command mxram 10000 proc Parsimony_matrix_SD6.tnt hold 10000000; outgroup 2; mult: replic 100000 hold 100; PIWE=12; mult; nelsen*; taxname= ; export* Parsimony_matrix_SD6_12KJB; RESAMPLE:jak replica9ons 1000[ mult = tbr replic 100 hold 10 ] from nelsen_tree_number; ##Replace nelsen_tree_number with the number of the nelsen tree## RESAMPLE:boot replica9ons 1000[ mult = tbr replic 100 hold 10 ] from nelsen_tree_number; ##Replace nelsen_tree_number with the number of the nelsen tree## Chronological scaling. The chronologically scaled phylogene9c tree can be reproduced by running the following command prompt into RStudio: (Note that, prior to running the below script, make sure that you have run script Bayesyan_matrix_and_script_SD4.nex [see above] and saved the resul9ng tree as "Setapedites.tree.txt".) setwd("C:/Users/ ") ##Set here the directory where the file has been saved## install.packages(c("geoscale", "strap"), dependencies=TRUE) install.packages("paleotree" , dependencies=TRUE) library(paleotree) library(strap) Setapedites.tree <- read.tree("Setapedites.tree.txt") Setapedites.ages <- read.table("Setapedites_age.txt", header=T) Setapedites.ts.tree <- DatePhylo(Setapedites.tree, Setapedites.ages) Setapedites.ts.tree$root.9me geoscalePhylo(Setapedites.ts.tree) pdf("Setapedites_tree_1.pdf", width=10, height=7) geoscalePhylo(Setapedites.ts.tree) dev.off() Setapedites.ts.tree <- DatePhylo(Setapedites.tree, Setapedites.ages, 2, "equal") pdf("Setapedites_tree_2.pdf", width=10, height=7) geoscalePhylo(Setapedites.ts.tree) dev.off() Setapedites.ts.tree <- 9mePaleoPhy(Setapedites.ts.tree, Setapedites.ages, "mbl", 4) pdf("Setapedites_tree_3.pdf", width=10, height=7) geoscalePhylo(Setapedites.ts.tree) dev.off() pdf("Setapedites_tree_4.pdf", width=10, height=7) geoscalePhylo(ladderize(Setapedites.ts.tree, right=FALSE), Setapedites.ages, cex.ts=0.5) dev.off() pdf("Setapedites_tree_last.pdf", width=28, height=12) geoscalePhylo(Setapedites.ts.tree, Setapedites.ages, cex.ts=1, cex.9p=1, cex.age=1, units=c("Period"), quat.rm=TRUE) dev.off()