Extended Data

Plant genes influence microbial hubs that shape beneficial leaf communities

Benjamin Brachi, Daniele Filiault, Paul Darme, Marine Le Mentec, Envel Kerdaffrec, Fernando Rabanal, Alison Anastasio, Matthew Box, Susan Duncan, Timothy Morton, Polina Novikova, Matthew Perisin, Takashi Tsuchimatsu, Roderick Woolley, Man Yu, Caroline Dean, Magnus Nordborg, Svante Holm, Joy Bergelson

correspondence to: [email protected]

Extended Data Fig. 1 to 7 Extended Table 1 to 3

1.0 Fungi ● / Helotiales / Helotiales family Incertae sedis ● Tremellomycetes / / 0.8 ● Tremellomycetes / / Tremellales family Incertae sedis ● unclassified / unclassified Ascomycota / unclassified Ascomycota ● Leotiomycetes / Helotiales / unclassified Helotiales ● / Pleosporales / unclassified Pleosporales ● Microbotryomycetes / Leucosporidiales / Leucosporidiaceae 0.6 ● Dothideomycetes / Capnodiales / Davidiellaceae ● Microbotryomycetes / Sporidiobolales / Sporidiobolales family Incertae sedis ● Tremellomycetes / unclassified Tremellomycetes / unclassified Tremellomycetes

0.4

0.2

0.0 S1_2012 S1_2013 S2_2012 S2_2013 N1_2012 N1_2013 N2_2012 N2_2013

1.0 Bacteria ● Betaproteobacteria / Burkholderiales / Oxalobacteraceae ● Alphaproteobacteria / Sphingomonadales / Sphingomonadaceae 0.8 ● Betaproteobacteria / Burkholderiales / Comamonadaceae ● Gammaproteobacteria / Pseudomonadales / Pseudomonadaceae ● Alphaproteobacteria / Rhizobiales / Rhizobiaceae ● Alphaproteobacteria / Rhizobiales / Aurantimonadaceae ● Flavobacteriia / Flavobacteriales / Flavobacteriaceae 0.6 ● Bacilli / Lactobacillales / Streptococcaceae ● Alphaproteobacteria / Rhizobiales / Methylobacteriaceae ● Alphaproteobacteria / Caulobacterales / Caulobacteraceae

0.4

0.2

0.0 S1_2012 S1_2013 S2_2012 S2_2013 N1_2012 N1_2013 N2_2012 N2_2013

Extended Data Fig. 1: Relative frequency of the first 10 most frequent OTUs. Each stacked bar (x-axis) corresponds to a location/year combination. The y-axis gives the proportion of the 10 most frequent OTUs. The colors correspond to the taxonomical assignments of OTUs given in the legend (class / / family).

Extended Data Fig. 2: Plants grown in different environments have different microbial communities. The plots represent the projection of each sample on the plane defined by the first two constrained components of the fungal (A and C) and bacterial (B and D) communities, describing variation among sites and year. The percentages in parentheses are the proportion of the total inertia (square root of the Bray-Curtis distance) explained by each component. In all plots each dot represents a sample. In A and B, colors indicate the site from which samples where collected. Experiments from the South are represented in red (S1) and yellow (S2), and experiments from the North in blue (N1) and dark blue (N2). In C and D, dark and lighter green identifies samples collected in 2012 and 2013, respectively. The barplots on the right represent the cumulative proportion of variance explained by the first 7 constrained components (CAP1 to 7) for the fungal and bacterial communities, respectively.

S1 2012 S1 2013

● ●

70 ● 70 ● 0.10 ● 0.10 ● 0.20 0.20 ● ● ● ● 0.08 ● 0.08 ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● 0.15 ● 0.15 ● ● ● ●● ● ● ● ● ● 0.06 ● 0.06 ● ● ● ● ● ● ● ● ● ● ● ● 50 50 ● ●● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ●● ● ● ●● ● 0.10 ● 0.10 ● ● ● ● ● ● ● ●● ● ● ● ● ● 0.04 ● 0.04 ● ● ● ● ● ● ● ●●● ● ● ●● ● ●● ● ● ●●● ● ● ●● ● ● ● ● ● ●● ●●●●● ● ● ● ● ●● ● ●●●● ●● ● ● ● ●● ● ● ● ● ● ● ●● ●●● ●● ●● ● ● ● ● ● ● ● ●●●● ● ●● ● ●●●●● ●● ●● ● ● ● ●● ● ● ● ●● repeatability ● repeatability ● ● ●● ● ●● ● ● ● ● ● ●● ●●● ● ● ●●● ● ●●● ●● ●● ●●● ●● ●● ● ●● ● ● 0.05 ●●● 0.05 ● ●● ●●● ● ●● ●● ●●●● ●● ● ●●● ● ● ● ● ●● ●● ●●●●●●● ● ● ●● ●● 0.02 ●●● 0.02 ●●●●● ●● ● ● ● ●●● ●●● ● ● ● ● ● ● ●●●● ●●●●●●● ●● ● ● ●●● ●● ● ● ● ●●● ● ● ●● ●● ●● ●●●● ●●● ●● ●● ●●● ● ●●● ● ● ●● ●●●● ●●●●●●● ● ●●● ●●● ●● ● ●● ● ● ● ●●●●● ●●● ●● ●● ●●●● ● ●● ●●●●●● ●●● ●● ●●● ●● ●●●●● ●● ●● ●●● ●●●●●● ●●●●●●●●●●●● ● ●●●●● ●●●● ●● ●●●●● ●●●●● ●●●●●●●●● ●●●●●●●●●●● ●●●●●● ●●● ●●●●● ● ● ● ●●● ●●●●●●● ●●●●●● ●●●●● ●●●● ●●● ●● ● ●●●●● ● ●●●● ●●●●●● ●●●●●●●● ●●● ●●● ●●●● ●●●● ● ●●●●●● ●●●●●●●●●●●● ●● ●● ● ●●● ●●● ● ●●●●●● ● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●● 30 0.00 0.00 30 0.00 0.00 F F B B 0.0 0.2 0.4 0.6 0.8 0.0 0.2 0.4 0.6 0.8

frequency (bins of 1%) quantiles frequency (bins of 1%) quantiles

10 % of OTUs with low CI 10 % of OTUs with low CI above 0.1%: 9.85% above 0.1%: 4.59% 0 0

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 repeatability repeatability

S2 2012 S2 2013

● ● 70 0.14 ● 0.14 70 0.10 ● 0.10 ● ● ● ● 0.12 0.12 ● ● ● ● ● ● ● 0.08 ● 0.08 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 0.10 0.10 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● 0.06 ● 0.06 ● ● ● ● ●● ● ●● ● ● ● ● ● 0.08 ● 0.08 ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ●● 50 ● 50 ● ● ● ● ● ● ● ● ● ●●● ● ● ● ● ●● ● ● ●●● ● ●● ● ●● ● ●● ●●● ● ● ●●● ● ● ● ●● ● ●● ● ● ● ●● ● ●● ●● ●● ● ●● ● 0.06 ● 0.06 ● ● ●● ●● ● ● ● ● ● ●● ●● ● 0.04 ● 0.04 ● ●● ● ● ● ● ●● ● ●●● ● ● ● ●● ●● ● ● ● ●● ●● ●● ● ●● ● ● ● ●● ● ●●● ● ● ●● ● ● ● ● ●● ●●●● ● ● ●● ●● ● ● ●●● ● ●●● ● ● ● ●●●●● ●● ● ● ●●● ●● ● ● ● ●● ● ●●● ●●● ●●● ●● ●● ●●●● ● ●●● ● ● ●●● 0.04 ●●●● 0.04 ● ●●● ●●● ●● ●● ● ● ● ●● ●● ● ● ●● ●●●●● ● ●●●● repeatability ●● ●● ● ● repeatability ●●● ● ● ● ● ● ● ●● ●●●● ● ●●● ● ● ● ●●● ●● ● ● ●● ●● ●● ●● ● ●●●●● ●●●●● ●●●● ●●●●● ●● ● ●●● ● ● ●●●● ● ● ● ●● ●●● ● ●●●●● ●●●● ● 0.02 ●● 0.02 ● ● ●●●●● ●●● ● ● ● ● ●●● ●●● ● ● ●●● ●● ●● ●● ● ●●●●● ●●● ●●● ●● ●●● ●● ● ● ●● ●●● ●●● ●● ●● ●●● ● ● ● ●● ●●● ●●●● ●● ●● ● 0.02 ●●● 0.02 ● ●● ● ●●●● ●●●● ●● ●●●● ●●●●● ● ● ● ● ●● ● ● ●●●● ●● ●●● ●●●●● ●●● ● ●● ●●● ●●● ● ●●●● ●●●● ●●●● ●●●●● ●● ● ●● ● ●●●● ●●●●● ● ●●●● ● ●● ● ● ●●● ●●●●●●● ●● ●●●●● ●● ●●● ●● ●● ●● ●●●●● ●●●● ●●●●●●●● ●●●● ● ●● ● ● ●●●● ●●●● ●●●● ●● ● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●● 30 0.00 0.00 30 0.00 0.00 F F B B 0.0 0.2 0.4 0.6 0.8 0.0 0.2 0.4 0.6 0.8

frequency (bins of 1%) quantiles frequency (bins of 1%) quantiles

10 % of OTUs with low CI 10 % of OTUs with low CI above 0.1%: 5.34% above 0.1%: 9.66% 0 0

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 repeatability repeatability

N1 2012 N1 2013

● ● 70 0.15 ● 0.15 70 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 0.20 0.20 ● ● ● ● ● ● ● ● ● ● 0.10 0.10 ● ● ● ● ● ● ● ●● ● ●● ●● ● ●● ● ●● 0.15 0.15 ● ●● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ●● ● ● ● ● 50 ● ● 50 ● ●● ● ● ● ● ● ●● ●●● ●● ●● ●●●●● ● ●● ●● ●●● ●● ● ● ● ● ● ●● ● ● ● ●● ● ●● ●●●● ● ● 0.10 ● 0.10 ● ●● ● ● ● ● ● ●● ● ● ● ●●● ●●●●● ●● ● ● ● ●●● ●●● ●● ● ● ●● ●● ●● ● ● ●● ● 0.05 ●●● 0.05 ●●● ●●● ● ● ● ● ● ●●● ● ● ●● ● ● ● ●● ●●● ●● ●●●●● ● ● ● ●●●● ●● ●● ● ● ● ●●●● ● ●● ● ● ●● ●● ●● ● ●●● ●●● ●●● ●● ● ●● ●● ●●●●●● ●● ●● repeatability ● repeatability ● ●●● ●●● ●●●● ●● ● ● ● ●● ●● ● ● ●●●● ●●● ●●●● ●● ● ●●● ● ●●● ●● ●●● ●● ●● ●● ● ●●● ● ● 0.05 ●●● 0.05 ● ●●●●● ●● ●●●●●●●●● ● ●● ● ● ● ●● ● ●● ●●●● ●●●● ●●● ●● ●●●●● ●●● ●●●●● ●●●●●●●●●● ●●● ●● ●●● ●● ● ●●●● ● ● ● ● ●●●●●● ●●●●● ● ●● ● ●● ● ●● ●●●●●●● ●●●●● ●●● ● ● ●●● ● ●● ●●● ●●●●●●●●●●● ● ●●●●● ●●● ●●●● ● ● ●●●●●● ●●●●●●● ●●●●●●●●● ●●●● ● ●● ●●●●● ●●●●● ●●●●●●●●●●● ●●●●●●●●●●●●● ●●● ● ●●● ● ●● ● ●●●●●●● ●●●●●●●●●● ●●●●●●●●●●●● ●●●●● ●●● ●● ● ●● ●●●●● ●●●●●●●● ●●● ●●●●●●● ●●● ● ● ● ●●● ●● ●●●●●● ●●● ●●●●●●●● ●●●●●●●●●●●●● ●●● ● ●●● ● ●●●●●●●● ●●●●●●● ●● ●●●●●●●●●● ● ●● ●● ● ● ● ●●●●●●●● ●●●●●●●●●●●● ●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●● 30 0.00 0.00 30 0.00 0.00 F F B B 0.0 0.2 0.4 0.6 0.8 0.0 0.2 0.4 0.6 0.8

frequency (bins of 1%) quantiles frequency (bins of 1%) quantiles

10 % of OTUs with low CI 10 % of OTUs with low CI above 0.1%: 12.5% above 0.1%: 7.38% 0 0

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 repeatability repeatability

N2 2012 N2 2013

● ●

70 ● 70 ● ● ●

0.30 0.30 ● 0.10 ● 0.10 ● ● ● ● ● 0.25 0.25 ● ● ● ● ●● 0.08 ● 0.08 ●●● ● ●● ● ● ● ● ●● ● ● ● 0.20 0.20 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● 50 50 0.06 ● 0.06 ● ● ● ● ● ● ● ● ● ● ● ● ● ● 0.15 0.15 ● ●●● ● ●● ● ● ● ● ● ●● ●● ● ● ● ●● ● ●● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●● ●●●●● ● ● ● 0.04 ●● 0.04 ●● ●● ● ●● ● ● ●● ●● ● ● ● ●● ● ● ● ● ● ●● ●● ● ●●●●● ● 0.10 ●● 0.10 ● ● ● ●●● ● ●● ●●●●●●●●● ●● ● ● ●● ● ●●● ● ● ● ● ● ● ● ●● ● ● ●● ● ● ● ●● ● ●● ●●●●●● ●●●● ●● repeatability ● ● ● ● repeatability ●●● ●● ●● ●●● ●● ●●●●● ● ● ● ●●● ●●●● ● ● ●● ●●●●● ●● ● ● ●● ●●●● ●●●● ●●●● ●●●● ● ●●●●●●● ● ● ●● ●●● ●●●●●●● ● ●●● ●●● ● ●●●● ● ●●●●●●● ●● ● ● ● ●●●● ●● ● ●● ● ●●●●● ●●● ● ●● ●●● ●● ● ●●●●● ● 0.02 ●●● 0.02 ●●●● ● ●● ●● ●● ●● ● ● ●● ● ●● ●● ●●● ● ●● 0.05 ●●● 0.05 ● ●● ● ●● ●●● ● ●●●●●●● ●● ●● ● ● ●●●● ●●● ●●●●●●● ●●● ● ●●●● ●● ● ●●● ●●●● ●● ●●●●● ●● ●●●●●● ●●●●●●●●●●● ●● ●●●●● ●●●●● ●● ●●●●●●●● ●●●●●●●●●●●●● ●●●●●●●●●● ●●●● ●● ● ● ● ●●●●● ●●●●●● ●●●●●● ●● ●●●● ●● ●● ●●●●● ● ● ●●●●● ●●●● ●●● ●●●●● ●●●●●●● ●●●● ●●●●●●● ●●●● ●●●●●●● ●●●●●●●●●●● ●●●●●●●● ●●● ●● ●●●● ● ●● ●●●●●●● ●●● ●●●●●●● ●●●●●● ●●●● ●● ● ●●●● ●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●● 30 0.00 0.00 30 0.00 0.00 F F B B 0.0 0.2 0.4 0.6 0.8 0.0 0.2 0.4 0.6 0.8

frequency (bins of 1%) quantiles frequency (bins of 1%) quantiles

10 % of OTUs with low CI 10 % of OTUs with low CI above 0.1%: 6.13% above 0.1%: 6.09% 0 0

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 repeatability repeatability Extended Data Fig. 3: Distribution of heritability values estimated for individual OTUs. Each frame presents the distribution of heritability in one year/location combination, and the two smaller graphics within each frame represent the empirical cumulative frequency distribution of the heritability distribution and the boxplot contrasting bacterial (B) and fungal (F) OTUs.

Extended Data Fig. 4: Genome wide association results for F60 in S2 2013. A. Genome-wide Manhattan plot. The y-axis represents the –log10 of the p-value. The x-axis is the position along the 5 chromosomes of A. thaliana, represented in different shades of blue. The points in yellow correspond to SNPs with –log(combined p-value) ≥ 8 and the red vertical line indicates the center of the focal region for the following frames. B. Candidate genes in a ~40kb focal region indicated by the red vertical line in A. The arrows indicate the 5’-3’ orientation of genes (vertical position is random). C. Associations in the ~20kB region around associated SNP. Axes are the same as in A. D. The x-axis is the same as in C. (coordinate in bp). The y-axis gives the minor allele frequency of all SNPs in the focal region. The segments in yellow correspond to the associated SNPs. E. Pattern of pairwise linkage disequilibrium in the focal region. The grey lines in the frame between D. and E. translate the coordinates from position in bp (in E. and above) to the SNP order.

Extended Data Fig. 5: Association peak 2, chromosome 2, between 12.99 and 13.01 Mbp. A. Genome-wide Manhattan plot. The y-axis represents the –log10 of the combined p-value for the 10 genetically correlated heritable hubs. The x-axis is the position along the 5 chromosomes of A. thaliana, represented in different shades of blue. The points in yellow correspond to SNPs with – log(combined p-value) ≥ 8 and the red vertical line indicates the center of the focal region for the following frames. B. Candidate genes in a ~40kb focal region indicated by the red vertical line in A. The arrows indicate the 5’-3’ orientation of genes (vertical position is random). C. Associations in the ~20kB region around associated SNP. Axes are the same as in A. D. The x-axis is the same as in C. (coordinate in bp). The y-axis gives the minor allele frequency of all SNPs in the focal region. The segments in yellow correspond to the associated SNPs. E. Pattern of pairwise linkage disequilibrium in the focal region. The grey lines in the frame between D. and E. translate the coordinates from position in bp (in E. and above) to the SNP order.

Extended Data Fig. 6: Association peak 6, chromosome 4, between 8.53 and 8.65 Mbp. See Extended Data Fig. 5.

S1 2012 S1 2013 var. expl. = 24.48% var. expl. = 22%

0.0030 0.0015 0.0025

0.0020 0.0010 0.0015

0.0005 0.0010 0.0005

0.0000 0.0000 F6 F1 F5 F8 F5 F1 F8 B1 B6 B3 pc3 pc1 pc2 pc3 pc2 pc1 F13 F23 F31 F25 F58 F63 F70 F68 F96 F37 F76 F67 F91 F68 F96 F70 F59 F19 B32 B17 B11 B13 B52 B53 B25 B37 B46 F284 F178 F286 F240 F161 F147 F156 F146 F143 F132 F501 F152 B321 B175 B229 B434 B176 B131 B140 B293 B358 B149 B137 B184 B339 B150 B167 B195 B489 B270 B107 B156 B133 B161 B338

S2 2012 S2 2013 var. expl. = −5.24% var. expl. = 28.04%

8e−04 0.0015

6e−04 0.0010 4e−04

0.0005 2e−04

0e+00 0.0000 F1 F5 F1 F6 F8 B6 B3 pc3 pc2 pc1 pc3 pc2 pc1 F21 F60 F85 F87 F55 F36 F96 F68 F12 B77 B41 B17 B49 B56 B11 B60 B42 B17 B12 B58 B31 B37 B45 B62 B26 B95 B30 B39 B13 F587 F103 F210 F137 F281 F267 F273 F408 F293 F160 F390 F143 F147 F694 F183 F217 B192 B328 B149 B227 B247 B142 B290 B281 B114 B180 B402 B117 B149 B133 B175 B126 B330 B220 B162 B144 B106

N1 2012 N1 2013 var. expl. = 11.37% var. expl. = 4.7%

5e−04 8e−04 4e−04 6e−04 3e−04 4e−04 2e−04

1e−04 2e−04

0e+00 0e+00 F5 F8 F1 F8 F7 B2 B7 B4 pc1 pc3 pc2 pc1 pc2 pc3 F99 F26 F76 F49 F18 F30 F15 F44 F10 F11 F36 F21 F10 F19 B56 B68 B52 B64 B61 B20 B18 B50 B24 B35 B21 B42 B55 B28 B54 B66 B11 B40 B70 B76 B52 B32 B24 B33 B23 B48 B30 B12 B60 B25 B54 B35 F202 F165 F814 F145 F188 F213 F189 F798 F781 F586 F184 F188 F291 F136 F263 F822 F203 F396 F676 F201 F410 F292 F598 B106 B128 B174 B432 B124 B278 B375 B531 B272 B178 B171 B261 B265 B277 B304 B235 B482 B407 B311 B328 B146 B423 B103 B420 B291 B267 B403 B160 B155 B114 B364 B184 B125 B165

N2 2012 N2 2013 var. expl. = −0.86% var. expl. = 6.71%

5e−04 8e−04

4e−04 6e−04 3e−04 4e−04 2e−04

2e−04 1e−04

0e+00 0e+00 F1 B3 B6 B4 pc3 pc2 pc1 pc1 pc2 pc3 F90 F35 F34 B34 B16 B21 B60 B40 B26 B24 B73 B65 B81 B54 B85 B12 B88 B24 B99 B71 B26 F402 F318 F412 F936 F599 F244 F179 F146 F163 F150 F410 F235 F734 F263 F291 F116 F240 F119 F219 B132 B133 B452 B145 B127 B168 B105 B291 B274 B420 B364 B302 B219 B115 B198 B424 B421 B382 B134 B106 B323 B197 B150 B182 B110 B247

Extended Data Fig. 7: Heritable hubs contribute to fecundity variation. For each location/year, we fitted random forests regression models using breeding values per accession for fecundity as dependent variables. The explanatory variables were breeding values for heritable OTUs (including heritable hubs) and the first 3 components of the pairwise genetic distance (1-Kinship) matrix. In each frame, the y-axis gives the variable importance. Variables include heritable OTUs (blue), heritable hubs (yellow), and the genetic distance components (red).

Extended Data Table 1: Host variation has subtle impact on overall community variation. The first 3 columns indicate the community, location and year for which the analysis was performed. Nh stands for the number of principal coordinate components with significant broad sense heritability estimates (95% confidence intervals not overlapping 0). A total of 10 components were computed for each community/location/year combination. “VE” indicates the total amount of microbial community variation captured by the first 10 components and “he” provides an estimated proportion of total variation explained by the identity of host accessions. The overall host effects reported in the main text reflect the distribution of VE*he in this table. N samples and N OTUs indicate the number of samples and OTUs analyzed in each community/location/year combination.

gene location year Nh VE he N samples N OTUs 16S ada 2012 7 55.185 1.244 570 279 16S ada 2013 5 51.78 1.593 912 338 16S ram 2012 7 57.683 3.826 589 256 16S ram 2013 7 45.852 2.115 1021 375 16S rat 2012 4 57.188 1.204 695 250 16S rat 2013 8 45.904 1.958 973 295 16S ull 2012 7 49.242 2.529 1051 321 16S ull 2013 6 41.472 1.002 982 303 Total number of bacterial OTU/location/year combinations 2417 ITS ada 2012 6 61.578 1.656 530 210 ITS ada 2013 5 54.646 0.607 889 320 ITS ram 2012 9 60.073 2.743 578 163 ITS ram 2013 7 42.543 1.379 996 425 ITS rat 2012 4 47.986 1.990 696 163 ITS rat 2013 9 49.498 2.650 981 233 ITS ull 2012 9 58.717 5.248 996 156 ITS ull 2013 3 56.253 0.724 990 244 Total number of fungal OTU/location/year combinations 1914

Extended Data Table 2: Hubs are enriched for interkingdom connections (edges). For each site (first column) and year (second column), the table presents the results from a χ2 testing for enrichment in interkingdom edges when considering all edges, or edges involving at least one hub. B_B, B_F, F_F give the number of edges between 2 bacterial OTUs, a bacterial and a fungal OTU, and 2 fungal OTUs, respectively.

site year edges B_B B_F F_F χ2 p-value FDR adj. p-value S1 2012 all 1148 134 556 107.325 2.00E-05 0.0002 with hubs 369 61 34 S1 2013 all 976 140 844 26.922 4.00E -05 0.0003 with hubs 132 44 190 S2 2012 all 981 137 484 41.968 2.00E -05 0.0002 with hubs 277 41 50 S2 2013 all 1039 157 873 25.536 2.00E -05 0.0002 with hubs 132 49 162 N1 2012 all 1212 157 637 71.122 2.00E -05 0.0002 with hubs 240 29 23 N1 2013 all 1427 288 1185 25.593 2.00E -05 0.0002 with hubs 271 104 295 N2 2012 all 951 71 430 15.604 6.20E -04 0.0050 with hubs 183 16 42 N2 2013 all 1720 229 1473 62.111 2.00E -05 0.0002 with hubs 254 94 341

Extended Data Table 3: Breeding values of heritable hub F8 have a significant positive effects on breeding-values for fecundity in linear mixed model including a random term accounting for IBS kinship (K) among accessions. For each location and year, models were defined as follow Yi~ hh1i + … + hhni + K + th th th errori, where Yi is the fecundity breeding value for the i accession, hhni are fixed effects for the breeding values for the i accessions for the n heritable hub identified in the location/year combination, K is a random term capturing genome wide relatedness among accessions (kinship) and errori captures residuals using a normal distribution. Fixed coefficients Random effects experiment sample size coefficients Value Std Error z p-value Group Std Dev Variance S1 2012 196 Intercept 0.00014 0.00049 0.29 7.70E-01 K 8.61E-03 7.41E-05 F8 0.00701 0.00157 4.48 7.60E-06 Residual error 0.006816532 S1 2013 198 Intercept 0.00022 0.00052 0.42 6.70E-01 K 8.79E-03 7.73E-05 B107 -0.00476 0.00596 -0.80 4.20E-01 Residual error 0.007283165 B25 0.00284 0.00744 0.38 7.00E-01 F5 -0.00499 0.00705 -0.71 4.80E-01 F8 0.01529 0.00428 3.57 3.60E-04 S2 2012 177 Int 0.00064 0.00055 1.16 2.50E-01 K 9.22E-03 8.49E-05 B41 -0.00657 0.00523 -1.26 2.10E-01 Residual error 0.006883981 S2 2013 197 Intercept 0.00007 0.00032 0.21 8.40E-01 K 4.49E-03 2.02E-05 B26 0.00789 0.00387 2.04 4.10E-02 Residual error 0.004501831 F5 -0.01309 0.00351 -3.73 1.90E-04 4.08E-03 1.66E-05 F60 0.01130 0.00369 3.06 2.20E-03 0.007389621 F8 0.00883 0.00394 2.24 2.50E-02 N1 2012 160 Intercept 0.00080 0.00060 1.33 1.80E-01 K 4.08E-03 1.66E-05 B28 0.00816 0.00692 1.18 2.40E-01 Residual error 0.007389621 B38 -0.00707 0.00585 -1.21 2.30E-01 F8 0.00387 0.00289 1.34 1.80E-01 N1 2013 198 Intercept 0.00031 0.00051 0.61 5.40E-01 K 8.75E-03 7.65E-05 F19 0.00590 0.00713 0.83 4.10E-01 Residual error 0.007104794 F60 -0.02173 0.00817 -2.66 7.80E-03 F8 0.00276 0.00532 0.52 6.00E-01 N2 2012 179 Intercept 0.00064 0.00056 1.15 2.50E-01 K 1.49E-04 2.22E-08 B26 0.00262 0.00457 0.57 5.70E-01 Residual error 0.007457504 N2 2013 197 Intercept 0.00005 0.00043 0.12 9.10E-01 K 6.44E-03 4.15E-05 B12 -0.00403 0.00534 -0.75 4.50E-01 Residual error 0.005948529 B26 0.01793 0.00914 1.96 5.00E-02 B99 0.00164 0.00358 0.46 6.50E-01 F150 -0.00073 0.00460 -0.16 8.70E-01