The experiment consisted of adding Trichoderma T22 spores to soil: the fungus was added to two treatments, one with native soil and one with sterilised soil mixed with the native soil at a ratio of 1:9, representing a disturbance event that decreases the overall abundance of species. Soil bioreactors were incubated for seven weeks, and genomic DNA was extracted from a set of samples every week (one sacrificed bioreactor per sample). DNA extraction has been optimised to obtain high quantities of pure, long DNA filaments. The quantity and quality of DNA extracted each week were compared, and extracts from weeks zero (pre-incubation), three, five and seven were used for Illumina barcoding of fungi and bacteria (three replicates each). Soil DNA from the seventh week of incubation was then used in a metagenomic sequencing experiment on the GridION® platform (ONT, Oxford, UK) based on the Rapid Sequencing Kit (ONT, Oxford, UK) without PCR step. Six flow cells were used, and in each one, a library consisting of 4 multiplexed samples (one for each condition of the experiment) was sequenced. The DNA raw reads were basecalled via the Oxford Nanopore Technology pipeline "Guppy" supplied with the GridION®, using a GPU accelerated server available at NHM. Early downstream analysis components such as barcoding/demultiplexing, adapter trimming and alignment were contained within Guppy. The microbial community analysis was performed with Kraken2 software, used with a default database, and OmicsBox (Blast2Go) software was used to assemble the reads and align them to genes. MEGAHIT was applied for the assembly, and Prodigal was run for (procaryotic) gene prediction. Pfam-scan and Eggnog mapper were also used to obtain some annotations.
The introduction of the fungus had a negative impact on the abundance of taxa of bacteria, but it stimulated the presence of others, metabolically or physically linked to the fungus. The results suggest that more than an impact on bacteria's overall biodiversity, the fungus has favoured some groups at the expense of others, even creating new food webs and trophic niches. It was also found that the fungus led to more significant differences in the community composition compared to the control when incubated for seven weeks in the soil with an initial lower biodiversity. Therefore the level of resilience of the soil microbial community to T22 introduction appeared to be linked to the initial soil biodiversity. It also emerged that certain taxa of bacteria behaved differently in the presence of the fungus, depending on whether there were competitors or not. Thus, the success and resilience of certain groups of bacteria appeared to depend on the competition for space and nutrients. The positive effects of the fungus on specific taxa, in the sense of a population increase, were also only appreciated in samples with a lower abundance of microorganisms. At the functional level, the most striking differences were between intact soil and soil artificially depleted of its initial biodiversity. A differential abundance of a few dozen gene families was observed in the bioreactors where the fungus had been added compared to the control.
