Data and code for: Stronger historical contingency facilitates ecological specializations: an example with avian carotenoid networks

DATA & CODE FILE OVERVIEW This repository features the data, code, and results for the paper \"Stronger historical contingency facilitates ecological specializations: an example with avian carotenoid networks\", which can be found TBD. Authors: Erin S. Morrison, Liberal Studies, New York University ([e-mail]) Caitlin M. Hill, Tucson Botanical Gardens Alexander V. Badyaev, Department of Ecology & Evolutionary Biology, University of Arizona ([e-mail]) The manuscript examines how the structure of a network of biochemical reaction facilitates the transitions between adaptive states by examining the diversification of carotenoid metabolic networks used to produce pigments in the plumage of birds in relation to transitions in species diets. This data repository consists of nine data files, five code scripts, and this README document, with the following data and code filenames and variables described below: ESM is responsible for collecting the data and writing the code. Analysis Pipeline: 1. SupplementaryCode_S1_ConsensusPhylogeny.txt is a Python code run in Unix to construct a majority rule ultrametric consensus phylogeny (SupplementaryData_S5_UltrametricConsensusTree.nex) from 1,000 trees randomly sampled from the Jetz et al. 2014 dataset on birdtree.org (Supplementary Data S4). 2. SupplementaryData_S5_UltrametricConsensusTree.nex is the majority rule ultrametric consensus tree that was used to calculate ancestral reconstructions for all carotenoid compounds and reactions using the Unix program r8s (see SupplementaryCode_S2_r8s.txt for an example of an ancestral reconstruction for the compound lutein). The ancestral reconstructions were based on the binary states of each compound and reaction at the tips of the phylogeny. This data is located in SupplementaryTable_S3_BinaryNetworks.csv. The results of the ancestral reconstructions are summarized in SupplementaryTable_S4_ancestralreconstruction.csv. 3. SupplementaryCode_S3_IndependentContrasts.R was used to calculate independent contrasts for the number of metabolically-degenerate compounds in species networks and the modified Shannon diversity index. The raw data for these measures is located in SupplementaryTable_S1_CarotenoidNetworkMeasures.csv and this code also uses SupplementaryData_S5_UltrametricConsensusTree.nex as the phylogeny for the calculation of the independent contrasts. The output of these analyses are located in SupplementaryTable_S2_IndependentContrasts.csv. 4. SupplementaryCode_S4_Phylogeny_Fig2_FigS1_.R was used to make the phylogeny figures for Figure 2 and Figure S1. The phylogeny used in the figures is SupplementaryData_S5_UltrametricConsensusTree.nex and the data displayed in Figure 2 is based on species network data located in SupplementaryTable_S1_CarotenoidNetworkMeasures.csv. 5. SupplementaryCode_S5_Figures3-5.txt is the SAS code for the calculations and statistics displayed in Figures 3-5. The raw data for this code is from SupplementaryTable_S1_CarotenoidNetworkMeasures.csv. The resulting calculations were graphed manually in SigmaPlot 10. Data Files: 1. SupplementaryTable_S1_CarotenoidNetworkMeasures.csv: Carotenoid network measures and diet categories of species under this study and all reconstructed ancestral networks. It includes the following variables: SpeciesName Species scientific name present in the phylogeny (with the exception of Colaptes auratus cafer). Ancestral networks correspond to internal node numbers in majority rule consensus phylogeny (Supplementary Figure 1). NodeN Number of compounds in species' network EdgesN Number of enzymatic reactions in species' network DietN Number of dietary compounds in species network (compounds acquired externally of network) DegenN Number of metabolically-degenerate compounds in species network (compounds derived from two or more biochemical pathways of similar lengths starting from more than one dietary compound) *Not identified for ancestral networks EdgeS Edge sensitivity (ξ) is the average of the fraction of compounds lost out of the total compounds in the network when a reaction is removed. When ξ = 0, the deletion a reaction has no effect on the production of any of the other compounds. When ξ = 1, the removal of any reaction causes the entire network to disappear. Networks that only contained dietary compounds (had no reactions to lose) were assigned an edge sensitivity score of 0. NodeS Node sensitivity (ξ) is the average of the fraction of compounds lost out of the total compounds in the network when a compound is removed. When ξ = 0, the deletion a reaction has no effect on the production of any of the other compounds. When ξ = 1, the removal of any reaction causes the entire network to disappear. LutS Fraction of compounds lost out of the total compounds in a species network when the dietary compound lutein is removed. Species without this dietary compound were assigned a value of 0 ZeaS Fraction of...