Dollo meets Bergmann: Morphological evolution in secondary aquatic mammals
Data files
Nov 01, 2024 version files 472.68 KB
-
Farina_et_al_compiled_data.csv
439.85 KB
-
Farina_et_al_scripts.zip
29.88 KB
-
README.md
2.96 KB
Abstract
Secondary transitions to aquatic environments are common among vertebrates, and aquatic lineages display several adaptations to this realm, some of which might make these transitions irreversible. At the same time, discussions about secondary transitions often focus only on the marine realm, comparing fully terrestrial with fully aquatic species. This, however, captures only a fraction of land-to-water transitions, and freshwater and semi-aquatic groups are often neglected in macroevolutionary studies. Here we use phylogenetic comparative methods to unravel the evolution of different levels of aquatic adaptations across all extant mammals, testing if aquatic adaptations are irreversible and if they are related to relative body mass changes. We found irreversible adaptations consistent with Dollo’s Law in lineages that rely strongly on aquatic environments, while weaker adaptations in semiaquatic lineages, which still allow efficient terrestrial movement, are reversible. In lineages transitioning to aquatic realms, including semiaquatic ones, we found a consistent trend toward an increased relative body mass and a significant association with a more carnivorous diet. We interpret these patterns as the result of thermoregulation constraints associated with the high thermal conductivity of water leading to a body mass increase consistent with Bergmann’s Rule and to a prevalence of more nutritious diets.
Description of the data and file structure
“Farina_et_al_compiled_data.csv” contains species information. “Name” column is species name; “Order_Name” and “Family_name” columns provide taxonomic information; “source_IUCN” and Source_phylacine” indicate from where the data came from. “Step” is the aquatic adaptation categories described in this study (A0: terrestrial; A1 and A2: semi-aquatic; A3: fully-aquatic); “Adaptation” column indicate which adaptation the species has (mainly applicable for A1 species for which active search was required) and “Reference” is from where the information came from; “Mass.g” is body mass data in grams (g); “Diet.Plant”, “Diet.Vertebrate” and “Diet.Invertebrate” columns give information about diet for each species (proportion).
Code/Farina_et_al_scripts
“Ancestral_reconstruction.R”: ancestral reconstruction analysis.
“FossilBM_var_mu_var_rate.R”: empirical FossilBM analysis with mammals phylogeny, body mass data, and aquatic classification. We used variable trends and variable rates across the phylogeny.
“FossilBM_HP_var_mu_var_rate.R”: empirical FossilBM analysis with mammals phylogeny, body mass data, and aquatic classification. HP means we added hyper-prior in the trend parameter. We used variable trends and variable rates across the phylogeny.
“sim_no_trends.R” and “sim_no_trends_HP.R”: FossilBM simulations with no trends, HP means we added hyper-prior in the mu0.
“sim_rnd_bg.r” and “sim_rnd_bg_HP.r”: FossilBM simulation with trends, but random evolution in the background. HP means we added hyper-prior in the mu0.
“sim_neutral_bg.r” and “sim_neutral_bg_HP.r”: FossilBM simulation with trends, but neutral evolution in the background. HP means we added hyper-prior in the mu0.
“simmap_count.R”: stochastic mapping using the mean best matrix from the ancestral reconstruction results.
“simmap_per_clade.R”: stochastic mapping for each order of mammals.
“calculate_effect_of_trends.R”: script used to calculate the relative change in body mass per million year
“getAncRecData.R”: script used to save the ancestral reconstruction results.
“calculate_error_sim.r”: calculate “no trend” simulation error.
“calculate_coverage_sim.R”: coverage for rnd and neutral bg simulations.
“Preparing_trees.R”: script used to drop, replace, compare and save que trees used in this study analyses.
“dtest_phytools.R”: script used to perform the D-test using previously made stochastic maps.
“fossilbm_var_mu_const_rate.R”: empirical FossilBM analysis using variable trends, but constant rates across the phylogeny.
“fossilbm_const_mu_var_rate.R”: empirical FossilBM analysis using using constant trend and variable rates.
“fossilbm_const_mu_const_rate.R”: empirical FossilBM analysis using using constant trend and constant rates.