https://doi.org/10.5061/dryad.0p2ngf277

This folder contains the dataset & analytical scripts of the manuscript entitled "Evolutionary lability of sexual selection and its implications for speciation and macroevolution".

Authors & institutions:

Matheus Januario[a] Renato C. Macedo-Rego[b,c], Daniel L. Rabosky[a]

Author affiliations:

a. Museum of Zoology & Department of Ecology and Evolutionary Biology, University of Michigan, 2032 Biological Sciences Building, Ann Arbor, MI, 48109, (USA)

b. Programa de Pós-Graduação em Ecologia, Universidade de São Paulo, Rua do Matão, 321, Travessa 14, São Paulo, SP, 05508-090 (Brazil)

c. Departamento de Biologia Geral, Universidade Federal de Viçosa, Av. Peter Henry Rolfs s/n, Viçosa, MG, 36570-000 (Brazil)

Description of the data and file structure

Responsible for collecting data: Renato C. Macedo-Rego

Responsible for WRITING CODE: Matheus Januario

List of data files from "Time lability decouples sexual selection intensity from the origins of vertebrate biodiversity"

data.csv: measurements at population level (selection indices)

species_data_full.csv: measurements at species level, full dataset (averaged selection indices, speciation rates, dichromatism data, size dimorphism indices)

species_data_zero_only: measurements at species level, zero-only dataset (averaged selection indices, speciation rates, dichromatism data, size dimorphism indices)

finalTree.tre: phylogenetic tree used in all analyses

SSD_vertebrates.tsv: sex-specific size data (used as raw data for sexual size dimorphism analyses)

Armenta_etal_2008_data.txt: Dataset containing dichromatism measurements for many species of birds (Armenta et al 2008 Journal of Experimental Biology 211:2423–2430).

Blomberg_etal_2003_appendix.csv: Estimates of phylogenetic signal “Blomberg’s K/kappa” (only vertebrate estimates estimates from the Appendix 6 of Blomberg et al 2003 Evolution 57 no. 4, pages 717-745)

all_vert_tips_lambda.csv: Cooney & Thomas (2020) tip-specific speciation rate dataset.

speciation_rates_taxonomy: dataset used in the power comparison across vertebrate families with less than 200 spp (see details in Supplementary text part IV)

Variables (columns) within data files:

data.csv:

[1] "obs_ID" = unique ID for each observation

[2] "study" = unique ID for each study incorporated in sampling

[3] "paper" = Reference in short format for each study incorporated in sampling

[4] "year" = Publication year of study incorporated in sampling

[5] "species_unmatched_with_tips" = species name matching original reference

[6] "species" = species name that match with phylogeny (finalTree.tre) tips. If different from what the original references says, it reflects either a taxonomical change (usually just name spelling, not an actual decision on species concept). If name does not match phylo tips, use equivalent name assigne din column "phylo_species"

[7] "sex" = Biological sex

[8] "group" = short code for vertebrate order

[9] "intrasex_competition" = short code indicating if we were sure the studied individuals faced intrassexual competition

[10] "includes_ms_0" = short code for certainty (= 1) or uncertainty/absence of information (= 0) regarding the reporting of zero-valued individuals

[11] "ms_determination" = If the authors somehow influenced (=1) or did not influenced (=0) the mating success (e.g. offering more males for a set of females).

[12] "Is" = opportunity for sexual selection value

[13] "ms_n" = Sample size for Is

[14] "ms_mean" = Mean mating success

[15] "ms_var" = Variance in mating success

[16] "ms_lnCVR" = Difference between sexes

[17] "ms_meaning" = Whether mating success was measured by counting the number of mating events or the number of sexual partners

[18] "ms_repeat" = If the information on mating success is repeated in another line.

[19] "ms_exclusion" = logical indicating if Is data should be excluded by any reason found after the data was collected (e.g. because it represents a repeated measurement)

[20] "reason1" = reason for exclusion of mating success

[21] "If" = opportunity for selection operationalized with fertilization

[22] "fs_n" = Sample size for If

[23] "fs_mean" = Mean fertilization success

[24] "fs_var" = Variance in fertilization success

[25] "fs_lnCVR" = Difference between sexes

[26] "fs_repeat" = If the information on fertilization success is repeated in another line.

[27] "fs_exclusion" = logical indicating if If data should be excluded by any reason found after the data was collected (e.g. because it represents a repeated measurement)

[28] "reason2" = reason for exclusion of fertilization success

[29] "I" = opportunity for selection operationalized with reproduction

[30] "rs_n" = Sample size for I

[31] "rs_mean" = Mean reproduction success

[32] "rs_var" = Variance in reproduction success

[33] "rs_lnCVR" = Difference between sexes

[34] "offspring_age" = short code for offspring development stage where I was measured

[35] "rs_exclusion" = logical indicating if I data should be excluded by any reason found after the data was collected (e.g. because it represents a repeated measurement)

[36] "reason3" = reason for exclusion of reproduction success

[37] "bg_ms_n" = sample size for calculation of mating-based Bateman gradient

[38] "bg_ms_slope" = best estimate of slope of mating-based Bateman gradient

[39] "bg_ms_slope_2.5" = Slope 2.5% CI of mating-based Bateman gradient

[40] "bg_ms_slope_97.5" = Slope 97.5% CI of mating-based Bateman gradient

[41] "bg_ms_intercept" = best estimate of intercept of mating-based Bateman gradient

[42] "bg_ms_intercept_2.5" = intercept 2.5% CI of mating-based Bateman gradient

[43] "bg_ms_intercept_97.5" = intercept 97.5% CI of mating-based Bateman gradient

[44] "bg_ms_residual_se" = residual standard error of mating-based Bateman gradient

[45] "bg_ms_r2" = Pearson's correlation coefficient for calculation of mating-based Bateman gradient

[46] "bg_ms_exclusion" = logical indicating if bg_ms_slope data should be excluded by any reason found after the data was collected (e.g. because it represents a repeated measurement)

[47] "reason4" = reason for exclusion of mating-based Bateman gradient

[48] "bg_fs_n" = sample size for calculation of fertilization-based Bateman gradient

[49] "bg_fs_slope" = best estimate of slope of fertilization-based Bateman gradient

[50] "bg_fs_slope_2.5" = intercept 2.5% CI of fertilization-based Bateman gradient

[51] "bg_fs_slope_97.5" = Slope 97.5% CI of fertilization-based Bateman gradient

[52] "bg_fs_intercept" = best estimate of intercept of fertilization-based Bateman gradient

[53] "bg_fs_intercept_2.5" = intercept 2.5% CI of fertilization-based Bateman gradient

[54] "bg_fs_intercept_97.5" = intercept 97.5% CI of fertilization-based Bateman gradient

[55] "bg_fs_residual_se" = residual standard error of fertilization-based Bateman gradient

[56] "bg_fs_r2" = Pearson's correlation coefficient for calculation of fertilization-based Bateman gradient

[57] "bg_fs_exclusion" = logical indicating if bg_ms_slope data should be excluded by any reason found after the data was collected (e.g. because it represents a repeated measurement)

[58] "reason5" = = reason for exclusion of mating-based Bateman gradient

[59] "obs" = relevant observations

[60] "BAMM" = species-specific phylogenetic speciation (tip) rate, calculated from BAMM (i.e., lambda_{BAMM})

[61] "DR" = species-specific phylogenetic speciation (tip) rate, measured by the DR statistic (i.e., lambda_{DR})

[62] "clade" = shortcode for vertebrate order

[63] "phylo_species" = species name that match with phylogeny (finalTree.tre) tips. If different from what column "species" says, it reflects the replacement of a species missing form the DNA-based tree by another closely related species

[64] "study_type" = shortocode for general way in which data was obtained

species_data_full.csv

[1] "sp"

[2] "clade" = shortcode for vertebrate order

[3] "sex" = Biological sex

[4] "Is_avg" = opportunity for sexual selection (species average, calculated from population_data.csv)

[5] "Is_var" = opportunity for sexual selection (species variance, calculated from population_data.csv)

[6] "Is_min" = opportunity for sexual selection (species minimum value, calculated from population_data.csv)

[7] "Is_max" = opportunity for sexual selection (species maximum value, calculated from population_data.csv)

[8] "Is_n" = opportunity for sexual selection (species sample size of index values, calculated from population_data.csv)

[9] "I_avg" = opportunity for selection operationalized with reproduction (species average, calculated from population_data.csv)

10] "I_var" = opportunity for selection operationalized with reproduction (species variance, calculated from population_data.csv)

[11] "I_min" = opportunity for selection operationalized with reproduction (species minimum value, calculated from population_data.csv)

[12] "I_max" = opportunity for selection operationalized with reproduction (species maximum value, calculated from population_data.csv)

[13] "I_n" = opportunity for selection operationalized with reproduction (species sample size of index values, calculated from population_data.csv)

[14] "If_avg" = opportunity for selection operationalized with reproduction (species average, calculated from population_data.csv)

[15] "If_var" = opportunity for selection operationalized with fertilization (species variance, calculated from population_data.csv)

[16] "If_min" = opportunity for selection operationalized with fertilization (species minimum value, calculated from population_data.csv)

[17] "If_max" = opportunity for selection operationalized with fertilization (species maximum value, calculated from population_data.csv)

[18] "If_n" = opportunity for selection operationalized with fertilization (species sample size of index values, calculated from population_data.csv)

[19] "Bgm_avg" = mating-based Bateman gradient (species average, calculated from population_data.csv)

[20] "Bgm_var" = mating-based Bateman gradient (species variance, calculated from population_data.csv)

[21] "Bgm_min" = mating-based Bateman gradient (species minimum, calculated from population_data.csv)

[22] "Bgm_max" = mating-based Bateman gradient (species maximum, calculated from population_data.csv)

[23] "Bgm_n" = mating-based Bateman gradient (species sample size of index values, calculated from population_data.csv)

[24] "Bgf_avg" = fertilization-based Bateman gradient (species average, calculated from population_data.csv)

[25] "Bgf_var" = fertilization-based Bateman gradient (species variance, calculated from population_data.csv)

[26] "Bgf_min" = fertilization-based Bateman gradient (species minimum, calculated from population_data.csv)

[27] "Bgf_max" = fertilization-based Bateman gradient (species maximum, calculated from population_data.csv)

[28] "Bgf_n" mating-based Bateman gradient (species sample size of index values, calculated from population_data.csv)

[29] "phylo_sp" = species name that match with phylogeny (finalTree.tre) tips. If different from what column "species" says, it reflects the replacement of a species missing form the DNA-based tree by another closely related species

[30] "DR" = species-specific phylogenetic speciation (tip) rate, measured by the DR statistic (i.e., lambda_{DR})

[31] "BAMM" = species-specific phylogenetic speciation (tip) rate, calculated from BAMM (i.e., lambda_{BAMM})

[32] "male_weight_g" = Male weight in grams

[33] "female_weight_g" = Female weight in grams

[34] "male_SVL_mm" = Male snout vent length in millimeters

[35] "female_SVL_mm" = Female snout vent length in millimeters

[36] "male_length_at_maturity_mm" = Male length at maturity in millimeters

[37] "female_length_at_maturity_mm" = Female length at maturity in millimeters

[38] "reference_dimo" = Long code for reference in sex-specific size

[39] "SSD_w" = Lovich & Gibbons sexual size dimorphism index calculated with weight

[40] "SSD_svl" = Lovich & Gibbons sexual size dimorphism index calculated with svl

[41] "SSD_len" = Lovich & Gibbons sexual size dimorphism index calculated with length at maturity

[42] "dichro_PCA" = Armenta et al 2008 sexual dichromatism index calculated with a PCA using data on plumage reflectance color

[43] "dichro_segment_classif" = Armenta et al 2008 sexual dichromatism index calculated with segment classification, using data on plumage reflectance color

[44] "dichro_col_discriminab" = Armenta et al 2008 sexual dichromatism index calculated with a discriminability model

species_data_zero_only.csv

Same columns/variables as species_data_full.csv, but species-level selection indices estimates are different

SSD_vertebrates.tsv

[1] "sp" = species name that match with phylogeny (finalTree.tre) tips. If different from what column "species" says, it reflects the replacement of a species missing form the DNA-based tree by another closely related species

[2] "clade" = shortcode for vertebrate order

[3] "male_weight_g" = Male weight in grams

[4] "female_weight_g" = Female weight in grams

[5] "male_SVL_mm" = Male weight in grams

[6] "female_SVL_mm" = Female weight in grams

[7] "male_length_at_maturity_mm" = Male length at maturity in millimeters

[8] "female_length_at_maturity_mm" = Female length at maturity in millimeters

[9] "reference_dimo" = = Long code for reference in sex-specific size

all_vert_tips_lambda.csv

[1] "species" = species name in Cooney & Thomas (2020) dataset.

[2] "bamm.mcc.tree" = species-specific phylogenetic speciation (tip) rate, calculated from BAMM (i.e., lambda_{BAMM})

[3] "dr.mcc.tree" = species-specific phylogenetic speciation (tip) rate, measured by the DR statistic (i.e., lambda_{DR})

[4] "clade" = shortcode for vertebrate order

Armenta_etal_2008_data.txt

[1] "species" = species name in Cooney & Thomas (2020) dataset.

[2] "PCA" = Armenta et al 2008 sexual dichromatism index calculated with a PCA using data on plumage reflectance color

[3] "Segment_classification" = Armenta et al 2008 sexual dichromatism index calculated with segment classification, using data on plumage reflectance color

[4] "Colour_discriminability" = Armenta et al 2008 sexual dichromatism index calculated with a discriminability model

Blomberg_etal_2003_appendix.csv

[1] "Vertebrate" = Indicator if clade contains only vertebrates (1) or not (0).

[2] "Group" = shortcode for vertebrate order.

[3] "trait" = short description of trait. Originally from Blombe0rg et al 2003.

[4] "Trait_type" = Class of trait (i.e. "Body size", "Morphology", "Physiology". "Life history", "Ecology", or "Behavior"). Originally from Blomberg et al 2003.

[5] "samp_size" = number of species in original data source.

[6] "p.value" = P-value for signifance, based on permutation test (see Blomberg et al 2003 for details).

[7] "estimate" = Value of "Blomberg's K" (= "Kappa").

[8] "Source" = Short name for reference of original data source (see Blomberg et al 2003 for full references).

speciation_rates_taxonomy.csv

[1] "bamm.mcc.tree" = species-specific phylogenetic speciation (tip) rate, calculated from BAMM (i.e., lambda_{BAMM})

[2] "mean.dr.full.tree" = species-specific phylogenetic speciation (tip) rate, measured by the DR statistic (i.e., lambda_{DR})

[3] "clade" = Factor telling if species is a bird, a fish, an amphibian, a mammal, or a squamate

[4] "Family" = Taxonomic family of the species, as assigned by the "taxize" R package

List of tables (output) files from "Time lability decouples sexual selection intensity from the origins of vertebrate biodiversity"

PGLS_lambda_full_dataset.csv: results from PGLS analysis among speciation and selection indices (full dataset). See main text or script "PGLS_lambda_full_dataset.R" for more details on calculation (this is also Supplementary table 1)

PGLS_lambda_zero_only.csv: results from PGLS analysis among speciation and selection indices (zero-only dataset). See main text or script "PGLS_lambda_full_dataset.R" for more details on calculation (this is also Supplementary table 2)

PGLS_traits_full_dataset.csv: results from PGLS analysis among proxy traits and selection indices (full dataset). See main text or script PGLS_traits_full_dataset.R" for more details on calculation (this is also Supplementary table 3)

PGLS_traits_zero_only.csv: results from PGLS analysis among proxy traits and selection indices (zero-only dataset). See main text or script PGLS_traits_full_dataset.R" for more details on calculation (this is also Supplementary table 4)

phylosig_charac_evo.csv: trait evolution modeling results & phylogenetic signal results (see script "phylosig_charac_evo_mods.R" for details). (this is also Supplementary table 5)

summary_stat_signif: Summary of significant analyses. (this is also Supplementary table 6)

tip_lipas.csv: result of LIPA analysis. See main text or script "phylosig_charac_evo_mods.R" for more details on calculations

PGLS_lambda_stepwiseMS.csv: Results of the automatic stepwise model selection between speciation rate and indices, modulated by clade. See main text or script "PGLS_lambda_stepwiseMS.R" for more details on calculations.

PGLS_trait_stepwiseMS.csv: Results of the automatic stepwise model selection between proxy traits and indices, modulated by clade. See main text or script "PGLS_trait_stepwiseMS.R" for more details on calculations.

power_vert_families: Results of the power comparison across vertebrate families with less than 200 spp (see details in Supplementary text part IV)

Variables (columns) within data files (see main text for key on the labels of indices & speciation rate shortcodes):

PGLS_lambda_full_dataset.csv OR PGLS_lambda_zero_only.csv:

[1] "clade" = shortcode for clade used in analysis. "All" means all vertebrate orders

[2] "lambda" = shortcode for type of phylogenneitc speciation rate tip values

[3] "Index" = Selection index name

[4] "sex" = biological sex

[5] "sample_size" = Number of species used in analysis

[6] "intercept" = intercept of PGLS

[7] "slope" = slope of PGLS

[8] "delta_AICc_alt_mod" = Personal codification for easily handling of Delta AICc among models. Note that for code/programing convenience this was calculated as AICc_{Non-NULL} - AICc_{NULL}, so negative values appear when the null model was selected

[9] "p_intercept" = p-value of (non-null) intercept (test if different from zero)

[10] "p_slope" = p-value of (non-null) PGLS slope (test if different from zero)

[11] "rsq" = R-squared of (non-null) PGLS

[12] "pwr1_rate" = "r" parameter of the model employed in power analysis I

[13] "pwr1_delay" = "t" ("threshold") parameter of the model employed in power analysis I

[14] "pwrQuant_0.025" = Empirical 2.5% quantiles of simulated correlation values (power analysis II)

[15] "pwrQuant_0.5" = Empirical 50% quantiles (median) of simulated correlation values (power analysis II)

[16] "pwrQuant_0.975" = Empirical 2.5% quantiles of simulated correlation values (power analysis II)

[17] "pwr_n_signif" = Number of simulated datasets which we found to be signfiicant in power analysis II

[18] pwr_n_reject" = number of simulations excluded from final calculations. This is checking column, and all values should be equal to zero.

PGLS_traits_full_dataset.csv OR PGLS_traits_zero_only.csv:

[1] "clade" = shortcode for clade used in analysis. "All" means all vertebrate orders

[2] "Trait" = shortcode for proxy trait used

[3] "Index" = Selection index name

[4] "sex" = biological sex

[5] "sample_size" = Number of species used in analysis

[6] "intercept" = intercept of PGLS

[7] "slope" = slope of PGLS

[8] "delta_AICc_alt_mod" = Personal codification for easily handling of Delta AICc among models. Note that for code/programing convenience this was calculated as AICc_{Non-NULL} - AICc_{NULL}, so negative values appear when the null model was selected

[9] "p_intercept" = p-value of (non-null) intercept (test if different from zero)

[10] "p_slope" = p-value of (non-null) PGLS slope (test if different from zero)

[11] "rsq" = R-squared of (non-null) PGLS

[12] "pwr1_rate" = "r" parameter of the model employed in power analysis I

[13] "pwr1_delay" = "t" ("threshold") parameter of the model employed in power analysis I

[14] "pwrQuant_0.025" = Empirical 2.5% quantiles of simulated correlation values (power analysis II)

[15] "pwrQuant_0.5" = Empirical 50% quantiles (median) of simulated correlation values (power analysis II)

[16] "pwrQuant_0.975" = Empirical 2.5% quantiles of simulated correlation values (power analysis II)

[17] "pwr_n_signif" = Number of simulated datasets which we found to be signfiicant in power analysis II

[18] pwr_n_reject" = number of simulations excluded from final calculations. This is checking column, and all values should be equal to zero.

phylosig_charac_evo.csv:

[1] "Index" = Selection index name

[2] "sex" = biological sex

[3] "phylosig" = shortcode for type of phylogenetic singla measured

[4] "estimate" = Estimate for phylogenetic signal. Note every type of measurement works in a different scale

[5] "p-value" = p-value of test for significant phylogenentic singal

[6] "samp_size" = Number of species used in analysis

summary_stat_signif.csv:

[1] "clade" = shortcode for clade used in analysis. "All" means all vertebrate orders

[2] "sex" = biological sex

[3] "Independent variable" = the independent variable used in the PGLS

[4] "Predicted variable" = the predictedt variable used in the PGLS

[5] "Dataset" = The dataset used as the source of data

[6] "sample_size" = Number of species used in analysis

[7] "Effect" = Polarity of the effect in the PGLS

tip_lipas.csv:

[1] "Lcl_lipa" = LIPA vale

[2] "sp" = species name

[3] "sex" = biological sex

[4] "var" = Selection index name

[5] "significant" = logical indicating if LIPA value significantly deviates from others

PGLS_lambda_stepwiseMS.csv:

[1] "lambda" = shortcode for type of phylogenneitc speciation rate tip values

[2] "Index" = Selection index name

[3] "sex" = biological sex

[4] "sample_size" = Number of species used in analysis

[5] "intercept" = intercept of PGLS

[6] "sel_mod" = Name of selected model. Options are Full (lambda ~ index * clade), Simple (lambda ~ index+clade), or Null (lambda ~ 1)

[7] "loglik" = Log-likelihood of the best (selected) model.

[8] "AIC" = Akaike Information Criteria of the best (selected) model.

[9] "p_intercept" = p-value of the test for significant deviations from zero in the intercept

[10] "rsq" = R-square of the model, as implemented in the R package "phylolm"

[11] "N_nonzeropars" = Number of parameters that significantly deviate from zero.

[12] "dataset" = Indicator for "full" or "Zero-only" datasets.

PGLS_trait_stepwiseMS.csv:

[1] "lambda" = shortcode for type of phylogenneitc speciation rate tip values

[2] "Index" = Selection index name

[3] "sex" = biological sex

[4] "sample_size" = Number of species used in analysis

[5] "intercept" = intercept of PGLS

[6] "sel_mod" = Name of selected model. Options are Full (lambda ~ index * clade), Simple (lambda ~ index+clade), or Null (lambda ~ 1)

[7] "loglik" = Log-likelihood of the best (selected) model.

[8] "AIC" = Akaike Information Criteria of the best (selected) model.

[9] "p_intercept" = p-value of the test for significant deviations from zero in the intercept

[10] "rsq" = R-square of the model, as implemented in the R package "phylolm"

[11] "N_nonzeropars" = Number of parameters that significantly deviate from zero.

[12] "dataset" = Indicator for "full" or "Zero-only" datasets.

power_vert_families.csv:

[1] "clade" = shortcode for clade used in analysis. "All" means all vertebrate orders

[2] "family" = taxonomic family

[3] "pwr_1_delay" = "t" ("threshold") parameter of the model employed in power analysis I

[4] "ntips" = Number of species in the phylogeny of the evaluated family

Missing data code: NA

List of scripts from "Time lability decouples sexual selection intensity from the origins of vertebrate biodiversity"

Symbols used:

Script names, what they do (#), and which files they need (>) or create (*):

prepare_dataset.R

\# removes negative bateman gradients

\# aggregates selection intensity values by species and sex

\# removes zero-valued species

\# log-transforms sex sel indexes and spec rates:

\# Reads data on sex size and calculate SSD using Lovich & Gibbons (1992) index

\# Reads data bird dichromatism

\# incorporates trait (dichromatism and dimorphism) into dataset

\> data.csv [ measurements of pop-level intensity of selection ]

\> finalTree.tre [phylogeny with all species included in this study]

\> SSD_vertebrates.tsv  [sexual size dimorphism primary data]

\> Armenta_etal_2008_data.txt    [sexual dichromatism data]

\* creates  "sp_melt" and "sp_val" objects. Updates "data" object

prepare_dataset_zero_included.R

\# same as make_dataset.R, BUT it only includes data from population which we are certain that zero-success individuals are reported



\* creates  "sp_melt_zero_included" and "sp_val_zero_included", and "data_zero_included" objects

PGLS_lambda_full_dataset.R

\# Runs PGLS analysis among selection indices and phylogenetic speciation rates, and run power analyses (Full dataset)



\> objects created by script "prepare_dataset.R"



\* tables/PGLS_lambda_full_dataset.csv

\* figs/supp_figs/supfig_12_power_analysis_plots_lambda_full.pdf

PGLS_lambda_zero_included.R

\# Runs PGLS analysis among selection indices and phylogenetic speciation rates, and run power analyses (Zero-only dataset)



\> objects created by script "prepare_dataset_zero_included.R"



\* tables/PGLS_lambda_zero_only.csv

\* figs/supp_figs/supfig_13_power_analysis_plots_lambda_zero_only.pdf

PGLS_traits_full_dataset.R

\# Runs PGLS analysis among selection indices and proxy traits, and run power analyses (Full dataset)



\> objects created by script "prepare_dataset.R"



\* tables/PGLS_traits_full_dataset.csv

\* figs/supp_figs/supfig_14_power_analysis_plots_traits_full.pdf

PGLS_lambda_zero_included.R

\# Runs PGLS analysis among selection indices and proxy traits, and run power analyses (Zero-only dataset)



\> objects created by script "prepare_dataset_zero_included.R"



\* tables/PGLS_traits_zero_only.csv

\* figs/supp_figs/supfig_15_power_analysis_plots_traits_zero_only.pdf

phylosig_charac_evo_mods.R

\# Calculates phylogenetic signal, LIPAs, and does model testing (between Brownian, Early-Burst, Ornstein-Uhlenbeck, and White Noise models) for indices, proxy traits, and for body size (i.e. the "positive control")



\> objects created by script "prepare_dataset.R"    

\*  tables/phylosig_charac_evo.csv

\* tables/tip_lipas.csv

PGLS_lambda_stepwiseMS.R

\# Runs the automatic stepwise model selection between speciation rate and indices, modulated by clade. See main text for more info, and see table "PGLS_lambda_stepwiseMS.csv" for results".

PGLS_trait_stepwiseMS.R

\# Runs the automatic stepwise model selection between proxy traits and indices, modulated by clade. See main text for more info, and see table "PGLS_trait_stepwiseMS.csv" for results".

Scripts related to particular figures read the appropriate dataset and generate the file, usually as .pdf and .png. Important: some details (e.g. color legends) are added later manually.

For further details:

Please send questions to januarioml.eco [at] gmail [dot] com.