Genus-level ubiquity masks amplicon sequence variant structure in Aspergillus, Penicillium and Fusarium from Vitellaria paradoxa soils across three Nigerian savanna zones

Published: 6 October 2026| Version 1 | DOI: 10.17632/mykyd5dgd7.1
Contributors:
Victoria Modupe Nwaokolo,
,
,
,

Description

This dataset contains the data and R code used in the manuscript "Genus-level ubiquity masks amplicon sequence variant structure in Aspergillus, Penicillium and Fusarium from Vitellaria paradoxa soils across three Nigerian savanna zones". Rhizosphere and paired non-rhizosphere soils were collected beneath Vitellaria paradoxa (shea) trees at six sites, two in each of the Guinea (Zaria), Sudan (Kano) and Derived (Saki) savanna zones of Nigeria. The resulting 12 composite samples were analysed by ITS2 metabarcoding (primers ITS3/ITS4; Illumina MiSeq, 2 × 300 bp). Reads were denoised with DADA2, yielding 929,746 reads and 2,613 amplicon sequence variants (ASVs). Taxonomy was assigned with QIIME 1.9.1/UCLUST against the UNITE release of 20 November 2016. Soil physicochemical properties were measured in triplicate at the Soil Laboratory, Forestry Research Institute of Nigeria, Ibadan: pH, organic carbon and matter, particle size, total N, exchangeable K, Na, Ca and Mg, available P, and Cu, Zn, Fe and Mn. The dataset includes: (1) the ASV count table (12 samples × 2,613 ASVs); (2) ASV sequences in FASTA format; (3) the taxonomic assignment of each ASV; (4) sample metadata giving savanna zone, soil compartment and site pairing; (5) soil properties as means and standard deviations of three analytical replicates; (6) R scripts that reproduce all analyses and figures, together with their output tables and figures. The analyses examine occupancy, zone and compartment restriction, and within-genus ASV composition of the three most abundant named genera, Aspergillus, Penicillium and Fusarium. Exact permutation tests respect the paired, site-level design. Although all three genera occurred in every sample, 50–61% of their ASVs were detected in only one sample. Savanna zone explained 25–35% of within-genus compositional variation. Raw sequence reads are available in the NCBI Sequence Read Archive under BioProject PRJNA1531940. A README file describes every file, column and unit, and explains how to rerun the analyses (R ≥ 4.3 with the vegan package; run scripts/run_all.R).

Files

Steps to reproduce

1.Sampling: In 2018, soil was collected beneath Vitellaria paradoxa trees in the Guinea (Zaria), Sudan (Minjibir) and Derived (Saki) savanna zones of Nigeria, two trees (sites) per zone. At each site, rhizosphere soil was collected in triplicate at 15 cm depth and a paired non-rhizosphere sample about 10 m from the tree; triplicates were pooled, giving 12 composite samples in six site pairs (Pair column, sample_metadata.csv). 2.Soil analysis (FRIN Soil Laboratory, Ibadan; triplicate determinations, mean and SD reported): samples air-dried and sieved to <2 mm (ISO 11464); particle size by ISO 11277; pH in 1:2 soil:water with a glass electrode (Thomas, 1996); organic carbon by Walkley-Black (Combs & Nathan, 1998), organic matter = OC x 1.724; total N by block digestion (Bremner, 1996); available P by Olsen extraction with ascorbic acid colorimetry (Kuo, 1996); exchangeable K, Na, Ca, Mg and Cu, Zn, Fe, Mn by 1 M ammonium acetate (pH 7.0) extraction and AAS (Buck Scientific 210 VGP). 3.DNA and sequencing: DNA extracted from 0.25 g soil with the PowerSoil DNA Isolation Kit (MO BIO/QIAGEN). The fungal ITS2 region was amplified with primers ITS3 (GCATCGATGAAGAACGCAGC) and ITS4 (TCCTCCGCTTATTGATATGC) and sequenced by Zymo Research on an Illumina MiSeq (2 x 300 bp). Raw reads: NCBI SRA BioProject PRJNA1531940. 4.Sequence processing: all samples denoised together with DADA2 (primer trimming, quality filtering, error modelling, chimera removal). Read pairs were truncated and joined with a 10-N spacer (129 nt + 10 N + 127 nt per ASV); ASVs with fewer than 11 reads per sample were removed. Output: 929,746 reads (44,553-115,117 per sample), 2,613 ASVs. Taxonomy assigned in QIIME 1.9.1 with UCLUST (similarity 0.90, max 5 hits) against UNITE QIIME release 20.11.2016 (97%). 5.Analysis (R 4.3.3, vegan 2.6-4): the community table was rarefied to 44,553 reads per sample (seed 20260928) and ASVs of Aspergillus, Penicillium and Fusarium extracted. Analyses: occupancy classes and zone/compartment restriction versus chance; genus relative abundance (paired signed-rank, Kruskal-Wallis on site means, split-plot ANOVA); within-genus composition (two-factor PERMANOVA on Bray-Curtis and Jaccard, dispersion tests, PCoA); Spearman (BH-corrected) and Mantel tests against soil. Permutation tests are exact and respect the paired design (90 zone relabellings among sites, 64 within-site compartment swaps). To reproduce: install R (4.3 or later) and vegan, unzip the dataset and run "Rscript scripts/run_all.R" from the top folder. All tables and figures are recreated in results/ in under a minute. README.md describes every file, column and unit.

Institutions

Categories

Metagenomics, Biodiversity of Soil Microorganisms, Rhizosphere Fungi

Licence