Additional file 1 of Destabilization of mutualistic interactions shapes the early heat stress response of the coral holobiont

Supplementary Material 1. Supplementary figures: Fig S1. Sample clustering for outlier detection in host and Symbiodiniaceae. Heat stress (no; low: < 1 DHW; moderate: 3 DHW) is defined here as ‘stress’. Only samples passing QC are included. Fig S2. Coral respiration rates throughout the heat stress experiment. Mean ± SE is shown over time, and DHW are reported for the heat treatment only (DHW = 0 under ambient conditions). Fig S3. Principal Component Analysis (PCA) of gene expression profiles in host and Symbiodiniaceae on transformed (variance stabilizing transformation) and centered data (left) and multilevel PCA on the same dataset (right) to account for the effect of colony of origin. Fig S4. ASV rarefaction curves of 16S rRNA gene sequences for coral (baseline, T0-T5), seawater and feed (rotifers and microalgae) samples. Fig S5. Non-metric multi-dimensional scaling (NMDS, sqrt-rooted data; stress = 0.17) based on Bray–Curtis dissimilarities calculated on relative abundance of ASVs in coral (Porites lutea), seawater and feed (rotifers and microalgae) samples. Fig S6. Mean relative abundance of dominant microbial families in seawater and feed (rotifers and microalgae) samples. For seawater, the microbial community structure is shown under ambient and heat treatment over time (T0, T3, T5). Fig S7. Changes in relative abundance of the dominant 10 microbial families (> 5% mean relative abundance) across treatments (ambient, heat) and time (baseline, T0—T5) in the coral samples throughout the heat stress experiment. Box = inter-quartile range (IQR), line in box = median, whiskers = minimum and maximum values not outliers (i.e. -/ + 1.5*IQR). Fig S8. Changes in mean relative abundance of differentially abundant ASVs between ambient and heat treatment over time (T0-T5). Differentially abundant ASVs were identified using DESeq (p < 0.01; adjusted post hoc tests), with the restriction that only ASVs present in at least 50% of the samples in one of the compared groups were considered. Significant differences are indicated with asterisks (heat vs ambient for the respective time point). Taxonomic assignment for each ASV is shown as Family, Genus, Species. Box = inter-quartile range (IQR), line in box = median, whiskers = minimum and maximum values not outliers (i.e. -/ + 1.5*IQR). Fig S9. NMDS (sqrt-rooted data; stress = 0.21) based on Bray–Curtis dissimilarities calculated on relative abundance of ASVs in coral samples during the heat stress experiment (baseline, ambient, heat) in the three colonies of origin (A, B, C). Fig S10. Shannon diversity index representing alpha diversity of the microbiome in the coral samples over time (baseline, T0-T5) and across treatments (ambient and heat). Box = inter-quartile range (IQR), line in box = median, whiskers = minimum and maximum values not outliers (i.e. -/ + 1.5*IQR). Fig S11. The most discriminative features of each omics-dataset along component 1 and 2 (based on DIABLO analyses), and their correlations. The most outer lines represent the abundance level of each variable under ambient, low stress and moderate stress. Features’ taxonomy / functional annotations are outlined in Table S17.