README

Code and settings for the analyses of the paper Incorporating evolutionary and threat processes into crop wild relatives conservation by Wolke Tobón-Niedfeldt, Alicia Mastretta-Yanes, Tania Urquiza-Haas, Bárbara Goettsch, Angela P. Cuervo-Robayo, Esmeralda Urquiza-Haas, M. Andrea Orjuela-R, Francisca Acevedo Gasman, Oswaldo Oliveros-Galindo, Caroline Burgeff, Diana Rivera-Rodríguez, José de Jesús Sánchez González, Jesús Alarcón-Guerrero, Araceli Aguilar-Meléndez, Flavio Aragón Cuevas, Valeria Alavez, Gabriel Alejandre-Iturbide, Carlos-H. Avendaño-Arrazate, César Azurdia Pérez, Alfonso Delgado-Salinas, Pablo Galán, Manuel González-Ledesma, Jesús Hernández-Ruíz, Francisco G. Lorea-Hernández, Rafael Lira Saade, Aarón Rodríguez, Dagoberto Rodríguez Delcid, José Ariel Ruiz-Corra, Juan José Santos Pérez, Ofelia Vargas-Ponce, Melania Vega, Ana Wegier, Martín Quintana-Camargo, José Sarukhán and Patricia Koleff.

The data used by the scripts below is provided in a separate Dryad repository (https://doi.org/10.5061/dryad.7m0cfxpxm). Scripts use relative paths assuming that the data is organised as explained below.

Dependencies

Analyses were carried out in: Zonation version 4 and in R version 3.5.1 and 3.6.09. The following R packages were used: ENMeval_0.3.0, purrr_0.3.4, tidyr_1.0.2 dplyr_1.0.2, ggplot2_3.3.3, readr_1.4.0, gridExtra_2.3, ggnewscale_0.4.5, scatterpie_0.1.5, pophelper_2.3.1, rgdal_1.4-8, raster_3.4-5, sp_1.4-4, scales_1.2.0 , rgl_0.107.10, vegan_2.6-2 and pcadapt_4.3.3. Additional R dependencies can be consulted in the html notebook of the r scripts (session info section). Maps were plotted with ArcGIS 10.2.2.

Analyses and Figures made in R

Code for analyses and plots made in R is available in the /bin directory of this repository in the .R and .Rmd scripts detailed below. For each .Rmd there is a .html version of it with an R notebook showing the code and output.

PCAdapt_PGD_teocintles.Rmd

Performs the PCA with the genetic data of Z. m. parviglumis.

Used to produce Figure 3c (3D PCA of genetic data).

Data used by this script:

• Genetic data for PCA: ../data/genetic/output/bytaxa_parviglumis.bed
• Samples metadata including PGD where they fell: ../data/genetic/output/PGD_points_parvi.txt"

Data available at the Dryad repository as: genetic.tar.gz.

plot_admixture_PGD_teocintles.Rmd

Explores admixture results to perform:

• Figure 3a (admixture plot) and Figure 3b (map with points).

• Supplementary Figure 5 (map with admixture pies).

Data used by this script:

• Zea mays parviglumis SDM: /data/spatial/modelosDarwinZea/Zea_mays_parviglumis.tif

• proxies of genetic differentiation croped for Z. m. parviglumis SDM:"../data/spatial/areasProxyDivGen/crop_to_sp/PGD_Zea_mays_parviglumis.tif"

• Z. m. parviglumis CV error (admixture output): "../data/genetic/output/admixture/parvi/Kerror_parviglumis.txt"

• Z. m. parviglumis Q files for K= 25 and K= 13 (admixture output): "../data/genetic/output/admixture/parvi/bytaxa_parviglumis.25.Q" and "../data/genetic/output/admixture/parvi/bytaxa_parviglumis.13.Q".

• Passport data for each sample: "../data/genetic/output/ADN_pasap_3604.txt"

• Fam file with sample list: "../data/genetic/output/bytaxa_parviglumis.fam"

Data available at the Dryad repository as: genetic.tar.gz, SDM_rasters.tar.gz and PGD_croptoSDM_rasters.tar.gz.

plot_scenarios_PGD_parvi.R

Plots the barplot of Figure 3d.

Data used by this script:

• Spatial output of zonation solution for 20% Mexico: /data/comparations_output/sol_prop_spp.txt

• Samples metadata including PGD where they fell: /data/genetic/output/PGD_points_parvi.txt

Data available at the Dryad repository as: Comparing_output_txt.zip and genetic.tar.gz.

The admixture plot, map, PCA and bar plot were joined using Inkscape to make:

• Figure 3. Genetic diversity of Zea mays subsp. parviglumis, a maize CWR, represented in the proxies of genetic differentiation (PGD).

plot_scenarios_PDG.R

Performs the plot an analyses:

• Figure 4. Performance of five systematic conservation planning scenarios to represent conservation features of Mesoamerican crop wild relatives, considering 20% of Mexico’s terrestrial area.

• Supplementary analyses suggested during review (only stats are mentioned in the ms text)

Data used by this script: /data/comparations_output/sols_summary_spp.txt produced by the script spatial_analyses_zonationVSproxiesdivgen_all_ms.R

Data available at the Dryad repository as: Comparing_output_txt.zip

plot_Zonation_curves.R

Performs the analyses and plots of the following figures:

• Figure 6. Performance curves quantifying the representation of proxies of genetic differentiation within the distribution of crop wild relatives in Mexico, based on the hierarchical landscape priority rank map.

• Supplementary Figure 9. Performance curves of how the proportion of taxa distribution ranges increased with the amount of land area.

Data used by this script:

• Read IUCN category per taxa: /data/spatial/Zonation_final_solutions/IUCN_threat_category.csv".

• Zonation curves output and features for all taxa: /data/spatial/Zonation_final_solutions/E_final_Todos.curves.txt and /data/spatial/Zonation_final_solutions/E_final_Todos.features_info.txt.

• Zonation curves output and features but for taxa exclusively distributing in natural vegetation: /data/spatial/Zonation_final_solutions/E_final_VegPyS.features_info.txt and /data/spatial/Zonation_final_solutions/E_final_VegPyS.curves.txt.

• Zonation curves output and features but for taxa associated to different habitats, i.e. natural vegetation, agricultural and urban areas: "../data/spatial/Zonation_final_solutions/E_final_HabVarios.features_info.txt" and "../data/spatial/Zonation_final_solutions/E_final_HabVarios.curves.txt".

Data available at the Dryad repository as: IUCN_threat_category.csv and Zonation_output_final.tar.gz

spatial_analyses_zonationVSproxiesdivgen_all_ms.R

For each taxon crops the proxies of gen div to the species distribution models. Then performs analyses to compare different ways to incorporate the PDG into the Zonation analyses.

Data used by this script:

• Zonation outputs and proxies of gen div:

/data/spatial/Zonation_output/01_MDP.rank.compressed.tif, # Zonation output with species only (n=116) #1
/data/spatial/Zonation_output/02_MDP_ZV.rank.compressed.tif, # Zonation output with species + LZ (n=143) #2
/data/spatial/Zonation_output/03_MDP_PDG.rank.compressed.tif, # Zonation output with species and PDG (n=218) #3
/data/spatial/Zonation_output/04_MDP_vs_PDG.rank.compressed.tif, # Zonation output with species vs PDG (n=5004) #4
/data/spatial/Zonation_output/05_MDP_PDG_ADMU.rank.compressed.tif, # Zonation output with species and PDG as ADMU (n=116+1) #5
/data/spatial/areasProxyDivGen/PDG.tif") #Proxies div gen 

• Species distribution models of each species: /data/spatial/modelos_darwin_all_final/*.tif

Input data available at the Dryad repository as: Zonation_output_final.tar.gz, PGD_rasters.tar.gz and SDM_rasters.tar.gz.

Output of this script:

• The rasters of proxies of genetic differentiation cropped for each taxa: /data/spatial/areasProxyDivGen/crop_to_sp/.

• Area of proxies of div gen for each zonation solution: /data/comparations_output/sol_tidy_spp.txt

• Proportion of proxi of div gen for each zolnation solution in relation to its area in the sp distribution: /data/comparations_output/sol_prop_spp.txt

• Diversity indexes and mean of proportion: /data/comparations_output/sols_summary_spp.txt

Output data available at the Dryad repository as: PGD_croptoSDM_rasters.tar.gz and Comparing_output_txt.zip

Zonation analyses

Zonation configuration files are in:/bin/Zonation_config_files.

File contents as follows:

• Batch files, contain the command line to call Zonation in the cluster: .sh files

• Run settings file, defines the analysis settings: E_final.dat

• Biodiversity feature files, listing the biodiversity layers: .spp files

• Condition file; listing the habitat layers: habitat_features.txt

• Groups file; links the biodiversity features to habitat features using habitat groups: habitat_group_features.txt

• Species of Special Interest file; lists observation records for species without potential distribution model: SSI_list.txt

• Zonation configuration file: zonation_config.txt.

Data needed for the Zonation analyses are available at the Dryad repository as: Zonation_input_conservationfeatures.tar.gz, Zonation_input_habitat.tar.gz and Zonation_input_spp_points.zip.

Note: the taxa Physalis ixocarpa and Solanum endiense are mentioned in the spp list, but where excluded (0 in the first column) from the final analyses because the first turned to be a synonym and the second was not evaluated by the ICUN.

Non-coding analyses and Figures

Figures 2, 5 and Supplementary Figures 1,2,4,8,10, 12, 13 and 14 were made in ArcMap with the spatial data and Zonation raster outputs available in the Dryad repository.

Species distribution modelling

Species distribution modelling was done by a previous study using the custom function MNE.R, made by Juan Barrios and available at https://github.com/AngelaCrow/MNE_ParientesSilvestresCultivos. See that repository for further details.

Data used:

• CWR ocurrence points in .csv. These were dowloaded for each taxon from the Sistema Nacional de Información sobre Biodiversidad de México (SNIB) and Global Biodiversity Information Facility and curated as explained in Methods.
• 19 bioclimatic variables form Worlclim Version 1.4 from 1.
• Terrestial ecoregions proposed by Olson et al. en el 2001
• Evapotranspiration, aridity index, annual radiation data from 2-3.
• Bare soil and cultivated areas from 4.

1. Hijmans, R. J., Cameron, S. E., Parra, J. L., Jones, P. G. & Jarvis, A. Very high resolution interpolated climate surfaces for global land areas. Int. J. Climatol. 25, 1965–1978 (2005).
2. Zomer, R. J., Trabucco, A., Van Straaten, O. & Bossio, D. A. Carbon, Land and Water: A Global Analysis of the Hydrologic Dimensions of Climate Change Mitigation through Afforestation/Reforestation. International Water Management Institute 101, (2006).
3. Trabucco, A. & Zomer, R. Global Aridity Index (Global-Aridity) and Global Potential Evapo-Transpiration (Global-PET) Geospatial Database. (2009).
4. Tuanmu, M. N. & Jetz, W. A global 1-km consensus land-cover product for biodiversity and ecosystem modelling. Glob. Ecol. Biogeogr. 23, 1031–1045 (2014).

Species distribution models can be downloaded from: http://www.conabio.gob.mx/informacion/gis/) under the category: Biodiversidad > Agrobiodiversidad y agroecosistemas > Parientes Silvestres de Cultivos. See Supplementary Data File 6 for direct download link for each SDM.