Fig 1.
Situation of New Caledonia in the world, and of Maré Island within this archipelago.
Location of the different sampling sites in Maré Island. SF is for Short Fallow (brown disk); LF is for Long Fallow (lightgreen disk) and F is for Forest (forest green disk). Map realised with the R package marmap.
Fig 2.
Bar plots of relative abundance (in per cent) of fungi phyla (A), their corresponding guilds (B), and of bacteria phyla (C) and their corresponding traits (D). Conditions legend: SF is for Short Fallow; LF is for Long Fallow and F is for Forest.
Fig 3.
Non-metric multidimensional scaling (NMDS) for fungi phyla (A) and bacteria phyla (B). SF is for Short Fallow (brown letters); LF is for Long Fallow (light green letters) and F is for Forest (forest green letters).
Fig 4.
Analysis of the inter-sample relationships for fungi (A) and bacteria (B) with the Anaconda R package. For each, at the left, a hierarchical clustering on Euclidean distances on logarithm-transformed ASVs abundance with average clustering method, at the right a Principal component analysis (PCA), and at the bottom, a sample-to-sample heatmap based on rlog transformation with trim on too low represented ASVs. SF is for Short Fallow (brown letters); LF is for Long Fallow (light green letters) and F is for Forest (forest green letters).
Fig 5.
MA plots (A and B) from the mean of normalised count by their respective log fold change, and Venn diagrams from ASVs (C and D) realised for fungi (A and C) and bacteria (B and D) with the Anaconda R package. SF is for Short Fallow; LF is for Long Fallow and F is for Forest.
Fig 6.
‘Targeted’ analysis results graph for fungi (A) and bacteria (‘b’) and Archaea (‘a’) (B) with the Anaconda R package. Results are in Log2FC for the significant ASVs according to the precision of the taxonomic rank: species (‘s’), genus (‘g’), family (‘f’), order (‘o’), class (‘c’), phyla (‘p’), or kingdom (‘k’). SF is for Short Fallow (brown); LF is for Long Fallow (lightgreen) and F is for Forest (forestgreen).
Fig 7.
‘Global’ analysis results graph for fungi with the Anaconda R package for the three comparisons.
SF is for Short Fallow; LF is for Long Fallow, and F is for Forest.
Fig 8.
Venn diagram from the ‘targeted’ analysis results graph for plantae for the three comparisons.
Values are corresponding fold changes. Conditional blue values are under-represented ASVs (negative fold change) and conditional red values are over-represented ASVs (positive fold change). SF is for Short Fallow; LF is for Long Fallow, and F is for Forest.
Fig 9.
Synthesis of soil environment components potentially involved in the accumulation of plant and human pathogens with the reduction of the fallowing period in traditional yams culture in Maré Island.
Detected increases and decreases of each component, via the approaches implemented in Anaconda, are represented by arrows within boxes on the left. Related probable positive and negative effects on soil augmentation of detrimental organisms are indicated, i.e., that may favour (+) or suppress (-) pathogens. Potential interactions between other parameters than pathogens are also shown (between A. spirorbis and ectomycorrhizal fungi). The absence of apparent abiotic effects is represented. An important aspect also represented is the potentiality of soil pathogen populations reaching a threshold that could lead to substantial plant and human infections.