Global Biogeography of Environmental Bacterial Metal Resistance

Raw report: metal_resistance_global_biogeography Feeds into: environment-embedding-geography, environmental-resistome, scale-dependent-mobile-element-associations, spatial-sampling-effort-confounding, study-batch-confounding-of-environmental-associations

Overview

This preliminary report maps the global distribution of environmental bacterial metal resistance using metagenome-assembled genomes (MAGs) extracted from MGnify and geospatial metadata retrieved through the ENA batch API. Data extraction, coordinate retrieval, and NB02 spatial analyses were completed; production figures and remaining validation analyses are pending. ₁

Key Findings

• The study extracted 260,652 environmental MAGs from MGnify, of which 30,497 remained after filtering host-associated biomes. ₁
• 22,356 MAGs had usable geospatial coordinates, representing 73.3% coordinate coverage of the environmental MAG set. ₁
• ENA batch API retrieval returned 24,511 sample records; 16,964 (69.2%) had valid latitude/longitude pairs, leaving a 30.8% spatial data gap in public metagenomic archives. ₁
• Among the 22,356 MAGs with coordinates, 2.8% carried at least one metal-resistance type. Metal resistance was therefore rare in the global environmental MAG pool and substantially lower than implied by the 30,497-MAG filtered set. ₁
• A 5° grid analysis included 289 cells with at least 5 MAGs. Using Fisher’s exact tests with Benjamini–Hochberg false discovery rate (BH FDR) correction, it identified 11 significant hotspots with OR>2 and q<0.05, and 3 coldspots with OR<0.5 and q<0.05. ₁
• The top hotspot was the Atacama/Andean region of Chile at lat=-25°, lon=-70°, with 21.8% prevalence, OR=9.83, q=7.6e-12, and n=101 MAGs (see atacama desert; figure: figures/fig_nb02_global_hotspot_map.png). Eastern and central USA clusters at lat=40°, lon=-80° and -90° had OR=7.9 and 6.3, while an East/Southeast Asia cluster at lat=25-30°, lon=105-120° had OR=4.4–5.9. ₁
• Biome-stratified Fisher’s exact tests against the global baseline, with BH FDR correction, found soil enriched for metal resistance at 5.8% prevalence, OR=5.05, and q=1.8e-82; rhizosphere prevalence was 1.9%, OR=0.66, and q=0.30 (NS); and marine prevalence was 1.2%, OR=0.20, and q=9.2e-80. ₁

| Biome ₁ | n MAGs | Prevalence | OR | q | |---|---:|---:|---:|---:| | Soil | 7,939 | 5.8% | 5.05 | 1.8e-82 | | Rhizosphere | 422 | 1.9% | 0.66 | 0.30 (NS) | | Marine | 13,995 | 1.2% | 0.20 | 9.2e-80 |

The report interprets soil as significantly enriched for metal resistance (5.8%, OR=5.05) and marine samples as significantly depleted (1.2%, OR=0.20). The rhizosphere result is a null result (not significant). These biome-specific values support environmental-resistome by showing soil enrichment and marine depletion in the coordinate-filtered environmental MAG dataset (figure: figures/fig_nb02_biome_prevalence.png). ₁

• The ENA coordinate file covered all areas represented in the MAG coordinate dataset: 0 of 532 MAG grid cells lacked ENA coordinate coverage. Thus, the 30.8% gap of 16,964/24,511 valid pairs is a per-sample gap rather than a geographic gap; it reflects samples without coordinates, not grid cells with no samples. ₁
• The drafted global map uses geopandas and matplotlib to distinguish metal-resistant MAGs, defined as those with n_metal_types > 0, from susceptible MAGs and to colour them by metal diversity. The map cell has a known NameError because import matplotlib.pyplot is missing from that cell. ₁
• The report states that metal-resistant genomes are not uniformly distributed among the 22,356 MAGs with coordinates, but that hotspot identification and sampling-effort correction are required before regional patterns can be interpreted. ₁
• The 2.8% global metal-resistance prevalence is much lower than the 21.8% prevalence of T4SS (type-iv-secretion-system, type IV secretion system horizontal-gene-transfer machinery) reported for comparison, supporting the interpretation that AMRFinderPlus metal-resistance genes and T4SS horizontal-gene-transfer machinery are distinct features that are not uniformly co-distributed. ₁

Caveats and Pending Analyses

• The report is preliminary. Its status section says that data extraction and coordinate retrieval were complete at NB01 and that NB02 spatial analysis and NB03 figures are pending. Later sections of the same report nevertheless present NB02 results and reference NB02 figures. Production figure completion remains pending. ₁
• The report’s interpretation section warns that, because of the 30.8% ENA coordinate gap, any global map of metal resistance based on public archives is subject to substantial geographic blind spots. It also names the question of which biomes are most underrepresented as a key open question. The later NB02 coverage check (0 of 532 MAG grid cells lacking ENA coordinate coverage) narrows this warning: the gap is per-sample, not missing coverage in the grid cells the data represent. Biome underrepresentation remains unresolved. ₁
• Expedition-level clustering remains untested. The report’s pending-analysis list proposes testing whether the Atacama cluster and the USA clusters are single-study artefacts by checking distinct sample_accession prefixes within each hotspot. That list gives the Atacama cluster as OR=9.8, while the NB02 hotspot results report OR=9.83. ₁
• Sampling-effort correction remains necessary; the report proposes testing whether hotspots persist after normalising by log(n_MAGs). ₁
• The matplotlib import must be fixed in the NB01 map cell before the production figure is generated. ₁

Slots Into

• environmental-resistome — global prevalence, biome enrichment and depletion, and spatial hotspots provide evidence about the geographic structure of environmental metal resistance. ₁
• environment-embedding-geography — coordinate coverage, the distinction between per-sample and geographic gaps, and pending sampling-effort correction identify limits on embedding resistance patterns geographically. ₁
• spatial-sampling-effort-confounding — the 73.3% coordinate coverage, the 30.8% per-sample ENA gap, the 5° grid hotspots and the untested log(n_MAGs) normalisation show that regional hotspots are not yet separated from sampling effort. ₁
• study-batch-confounding-of-environmental-associations — the Atacama and USA hotspots may be single-study artefacts, pending a check of distinct sample_accession prefixes per hotspot. ₁
• scale-dependent-mobile-element-associations — the contrast between 2.8% metal-resistance prevalence and 21.8% T4SS prevalence indicates that AMRFinderPlus metal-resistance genes and T4SS machinery are not uniformly co-distributed. ₁