Predictive functional converge after nitrate input
In FAPROTAX, most OTUs (> 90.7%) failed to associate with either N- or S-metabolism. In the absence of favorable electron acceptors, control communities were rich in methanogens (Figure S5A), which were archaea in the phylum Euryarchaeota . The number of putative nitrate reducers (e.g., members of genus Nitrospira ,Bacillus, Paracoccus ) and sulfate reducers (e.g., members of genus Syntrophobacter , Desulfococcus ,Desulfobulbus , Desulfovibrio ) were comparable (Figure S5B, S5C). For nitrate communities, the number of putative nitrate reducers kept decreasing, which contradicted the drastic nitrate reduction possibly because the actual nitrate reducers were yet to be collected in the FAPROTAX database. Members of genus Thiobacillus ,Paracoccus , Sulfuricurvum or Acidithiobacillus were putative AVS oxidizers and Thiobacillus in particular, was assigned to dark-sulfide-oxidation and dark-oxidation-of-sulfur-compounds.
In PICRUSt, we focused on the changes of gene copy number based on a normalized PICRUSt table (KEGG Orthology Level 2 and 3). Each sample was rarefied to an equal sum of predicted gene copies so that the copy number of each functional item represented its strength. When analyzing the general metabolism (ko09100, Level 2), the late nitrate communities (16-N and 32-N) exhibited the greatest difference from the rest. Five orthologues (ko09102, 09104, 09106-09108) were enriched and four (ko09101, 09109-09111) were suppressed (Figure 4). Within ko09102 (energy metabolism), nitrogen metabolism had higher background abundances (~25000 copies per community) than sulfur metabolism (~10500) and were more enriched after nitrate amendment (Figure S5E, S5F).