Data and scripts for "Effects of connectivity history on fungal community dynamics and recovery"

Data and scripts for reproducing the analyses, "Effects of connectivity history on fungal community dynamics and recovery". The input data consists of the following file, "Data_tables_clean.Rdata". The file is in R format. The R files need to be loaded using the load () function in R. "Data_tables_clean.Rdata" includes a total of 5 input data matrices: (1) Meta_logs, metadata including the environmental covariates both at the log and site level. The file contains 24 columns which are: 1) "Site_name_original", 2) "Log_ID", 3) "Site_name_translocation", 4)"Tree of origin_ID", 5) "Tree_sp", 6) "Inoculation", 7) "Diameter_cm" , 8) "Status", 9) "sample", 10) "Sampling_year", 11) "Site_type_original", 12) "Site_type_translocation", 13)"Latitude_original", 14) "Longitude_original", 15) "Latitude_trans", 16) "Longitude_trans", 17) "L.200_orig", 18) "L.500_orig", 19) "L.1000_orig", 20) "L.200_trans", 21) "L.500_trans", 22) "L.1000_trans", 23) "Translocation_type", 24) "Bark_thickness_cm". (2) otu.table.plausible.clean, which is a sample x OTU table including the number of reads for each OTU for the plausible OTU taxonomic identification. (3) otu.table.plausible.clean.w, which is a sample x OTU table including the relative read counts for each OTU for the plausible OTU taxonomic identification. (4) read.counts.plausible.clean, which includes the read counts per sample during the different phases of the bioinformatics pipeline for the plausible OTU taxonomic identification. (5) taxonomy.plausible.cl, which contains the taxonomic information at all taxonomic levels (i.e., from species to phylum) of the identified OTUs. The statistical analyses use joint species distribution modelling with the Hmsc package and non-metric multidimensional scaling (NMDS) with the vegan package. Venn diagrams (eulerr package) and boxplots (ggplot2 package) are also used for visual analysis. To perform the HMSC analysis, scripts Scr6, Scr6b, Scr6c, Scr7, Scr8, Scr9, Scr10, and Scr11 must be run consecutively. Scr6 defines the presence-absence models using 2024 and 2025 data, with translocation type as an explanatory variable, along with diameter, bark thickness, inoculation status, and sequencing depth. Scr6b defines the species richness models with the same data. Scr6c defines both species richness and presence-absence models using 2023 data; here, site type replaces translocation type, since the logs have not yet been moved. This results in 12 models (6 pine and 6 spruce) with a similar set of explanatory variables, except for translocation type and site type. All these models are fitted in the study. Scr7 imports and exports the Hmsc models for fitting on a cluster computer using Hmsc HPC. The bash script used for fitting the models in the cluster computer is included called "hpc_thin100.sh". Scr8 evaluates the convergence of the MCMC chains. Scr9 calculates and plots the contrasts for the beta plots of the species richness and presence-absence models from 2024 and 2025 and calculates confidence intervals for the beta parameters. Scr10 shows parameter estimates from the fitted models, in particular, beta parameters and the variance partitioning across environmental covariates. The remaining scripts, Scr1–Scr5, produce the plots shown in the study, including Venn, NMDS, and box plots.