Differentially abundant taxa in microeukaryote community
Based on the consistency between number of OUT_S and SH_99, we conclude that the OTU_S dataset provides a better estimate of total species richness. We thus used this dataset for further analysis of differences in communities associated with contrasting soil conditions. Across the total microbial eukaryotic community, 282 fungal and 383 protist OTU_Ss were present in both wet and mesic-dry conditions. A total of 195 fungal and 243 protists OTU_Ss were presented only in wet conditions, while 292 fungal and 299 protists OTU_Ss were detected only in mesic-dry condition (Fig. S13). Based on the DESeq analysis, only 15 were significantly differentially abundant across all OTU_S (Fig. S14). Four out of fifteen were protists, one of them belongs to genusPolymyxa and is only found in mesic-dry condition while the other three are found predominantly in wet soil condition (Supplementary datafile 4). Taxa in at least two of the Polymyxa genus have been reported as plant root endoparasites (Decroƫs et al., 2019; Neuhauser et al., 2014). Of the eleven OTU_S belonging to the fungal kingdom, only one was significantly more abundant in the mesic-dry condition. Five OTU_Ss belong to Ascomycota were identified until genus level (threeCistella , Psedeurotium and Stagonospora ), and two OTU_Ss belong to Basidiomycota, recognized to the order level. All these taxa were significantly more abundant in the wet condition (Supplementary datafile 4). Detection of significant association with the contrasting soil conditions is limited by the current sampling design with only ten samples from wet and mesic-dry, respectively. Additional samples would have captured more of the local community. In relation to the presence/absence of F. meleagris , 160 fungal and 231 protists OTU_Ss were detected only in F. meleagris samples, while 144 fungal and 159 protist OTU_Ss were observed in samples without F meleagris . In total, 465 fungal and 544 protist OTU_Ss were observed in both with and without F meleagris samples (Fig. S15). Taking into account the low number of samples no OTU_Ss were differentially abundant based on the DESeq analysis. This is likely a result of sampling large soil volumes with multiple microhabitats, where only some are affected by the target plant species. Our attempt to also sequence root associated communities, which would be expected to better capture specifically host plant associated microorganisms, failed due to low success rate of microeukaryote amplification from F. meleagris root samples (data not shown).