Functional, interspecific, and intraspecific diversity jointly shape growth and recovery of Synechococcus communities under environmental stress

Biodiversity-growth-recovery.R Purpose: Analyses how initial and assembled functional diversity (FD) predict community growth and recovery in microbial communities exposed to environmental stressors (atrazine, heating, or both). Data Reads all-data-with-diversity.rds, a multi-timepoint dataset of microbial communities varying in strain and species richness across four treatments (Control, Atrazine, Heating, Heating + Atrazine). Analysis is restricted to days 0, 10, and 20. Key variables computed change — log ratio of population/diversity between day 0 and day 10 (growth phase) recovery — log ratio between day 10 and day 20 (recovery phase) strain_richness / species_richness — parsed from strain/species combination strings Models Four GLS models with treatment-level variance heterogeneity (varIdent): Model Response Predictor of interest m1 Community growth Initial FD (diversity_0) m2 Community recovery Assembled FD (diversity_10) m3 FD change Initial FD (diversity_0) m4 FD recovery Assembled FD (diversity_10) All models include a diversity × species_richness × treat interaction and strain_richness as an additive covariate. Outputs final_results.xlsx — tidy coefficient table with significance stars for all four models Figure 1 (g1/g2) — scatterplots of initial/assembled FD vs community growth/recovery, faceted by treatment × species richness Figure 2 (g3/g4) — same for FD change/recovery Figure 3 — slope plots (make_slope_plot) showing per-treatment FD effect sizes with 95% CIs, assembled via patchwork Supplementary S2 — boxplot of functional diversity over time by strain richness, treatment, and species richness Diagnostic panels — residuals vs fitted, Q–Q, and histogram for each model Dependencies nlme, emmeans, broom, broom.mixed, tidyverse, ggplot2, ggpubr, ggstats, patchwork, writexl, openxlsx, scales