What this entity is
Canonical name: Benjamini–Hochberg false discovery rate correction. 1
Known aliases: BH FDR; Benjamini–Hochberg FDR. 1
Stable external identifier: None is specified in the source document. 1
This method corrects multiple-testing results by producing q-values for spatial and biome-stratified Fisher’s exact tests. 1 Its use in community metabolic ecology extends this role to pathway–metabolite correlations and ecosystem-comparison tests. 2 Its application to PGP gene co-occurrence and environmental-enrichment tests further supports its cross-domain use for controlling discovery claims across many gene pairs and ecological comparisons. 3 Application to phage-defense syndromes further supports this cross-domain use: 1,000 phylum-stratified column-permutation tests were corrected with BH-FDR, and 27 of 28 defense-system pairs were significant at q < 0.05. 4
Application to soil-metal functional genomics extends this record to high-throughput COG–metal associations: 2,355 associations were reported at FDR < 0.05 among 3,915 implied tests across nine metals and 435 COGs, a 60% discovery rate. 5 The report cautions that co-varying metals make tests non-independent, so BH-FDR may be anti-conservative under positive correlation and the true FDR may be higher than reported. 5
Use in metal-resistance biogeography
The study applied Fisher’s exact tests with Benjamini–Hochberg false discovery rate correction to a 5° grid containing 289 cells with at least 5 MAGs. 1 The corrected analysis identified 11 significant hotspots meeting OR>2 and q<0.05, plus 3 coldspots meeting OR<0.5 and q<0.05. 1 The top hotspot, in the Atacama/Andean region of Chile at lat=-25°, lon=-70°, had 21.8% prevalence, OR=9.83, q=7.6e-12, and n=101 MAGs. 1
The study also used the correction for biome-stratified tests against the global baseline. 1 For soil, the corrected result was 5.8% prevalence, OR=5.05, and q=1.8e-82; for rhizosphere, 1.9% prevalence, OR=0.66, and q=0.30 (NS); and for marine samples, 1.2% prevalence, OR=0.20, and q=9.2e-80. 1 These results support environmental-resistome by quantifying evidence for soil enrichment and marine depletion of metal resistance in the coordinate-filtered environmental MAG dataset. 1
The soil-metal analysis refines this environmental-resistome record: chromium and lead produced the strongest signals among nine metals, with transporters including ABC and RND systems and biosynthesis genes among the top hits. 5 However, the associations are observational, and co-contamination among chromium, copper, lead, and zinc leaves metal-specificity unresolved pending partial-correlation analysis. 5
Use in community metabolic ecology
The NMDC × pangenome analysis supports the method’s cross-domain applicability: BH correction identified leucine and arginine as the only two tested pathway–metabolite associations significant at q < 0.05, with leucine r = −0.390, q = 0.022, n = 62 and arginine r = −0.297, q = 0.049, n = 80. 2 The same correction found significant ecosystem differences for 17 of 18 amino-acid pathways; glycine, asparagine, and cysteine had q = 1.2×10⁻¹⁹, q = 3.8×10⁻¹⁴, and q = 1.5×10⁻¹³, respectively, while tyrosine was not significant (q = 0.71). 2 These results refine its established use in spatial and biome-stratified enrichment testing by covering both correlation-level and omnibus ecosystem comparisons. 2
Use in PGP pangenome ecology
In the PGP pangenome analysis, BH-FDR correction found 8 of 10 focal-gene pairs significant: five positive and three negative associations. 3 The strongest positive association was pqqC × acdS (OR = 7.24, n = 286 co-occurring species, q = 1.2e-83); nifH was negatively associated with hcnC (OR = 0.23, q = 5.8e-29) and pqqC (OR = 0.57, q = 2.9e-19). 3 This supports using the correction to identify robust gene-gene ecological associations rather than treating unadjusted co-occurrences as discoveries.
The correction also supported soil/rhizosphere enrichment claims: acdS had OR = 7.02, q = 5.1e-62, pqqC had OR = 2.90, q = 2.8e-53, and hcnC had OR = 1.85, q = 6.1e-08; nifH was depleted (OR = 0.60, q = 5.5e-08). 3 These results refine the prior record of corrected biome comparisons by extending it from metal-resistance prevalence to gene-trait distribution across pangenomes. 3
Use in phage-defense and prophage ecology
The phage-defense study used BH-FDR after 1,000 phylum-stratified column-permutation tests to assess 28 pairwise defense-system co-occurrences. 4 Twenty-seven of 28 pairs showed significant positive co-occurrence at q < 0.05; the strongest was R-M Type II × Gabija, with z = 46.1 and OR = 24.0. 4 CRISPR-Cas × CBASS was the only nonsignificant pair (z = 0.21, p_emp = 0.98). 4 This refines the method’s existing record by demonstrating its use with empirical permutation nulls and phylum-stratified syndrome tests, while the report cautions that permissive CRISPR-Cas detection may reduce syndrome specificity. 4
Application to prophage ecology supports this broader use in module-level environmental testing: constrained permutations preserving host-family and genome-size-quartile strata identified 8 significant module-by-environment enrichments at FDR < 0.05, including tail, head morphogenesis, and anti-defense enrichment in human-associated environments. 6 BH-FDR also refines the interpretation by showing that 57 module–abiotic correlations in 6,365 NMDC metagenomic samples passed FDR < 0.05, whereas no individual TerL lineage was environment-enriched after correction in 0/500 tests. 6 Together with the defense-syndrome results, these applications support phage-defense-syndromes-and-arms-race as a cross-domain setting for distinguishing reproducible phage-associated patterns from multiple-testing artifacts. 4,6
Use in Pseudomonas carbon ecology
Application to standardized GapMind carbon-pathway comparisons supports the method’s use for correcting many pathway-level ecological tests: among 7 Pseudomonas s.s. species and 189 Pseudomonas_E species, 43 of 62 pathways differed significantly by Mann–Whitney U testing after Benjamini–Hochberg correction at q < 0.05. 7 This refines the existing record from gene-pair, ecosystem, and enrichment tests to comparative metabolic-pathway completeness across clades. 7
The corrected comparison supports pathway-loss claims rather than treating the numerous pathway differences as unadjusted discoveries; the largest reported contrasts included xylose completeness of 0.0% versus 74.4%, ribose 27.9% versus 92.0%, and arabinose 0.0% versus 62.6% between the P. aeruginosa and P. fluorescens groups. 7 The same analysis found 19 pathways with no significant subgenus difference, showing that correction also preserves a distinction between broadly conserved pathways and pathway-specific contrasts. 7 A 999-permutation environment-association test additionally reported p = 0.006 for carbon profiles among free-living and plant-associated species, while a Random Forest achieved balanced accuracy of 0.408 +/- 0.169 against a 0.250 chance baseline. 7 These results support a modest ecological signal but refine interpretation by showing that statistically detectable pathway differences do not make carbon profiles sufficient for fine-grained environment prediction. 7
Interpretation and limitations in soil-metal genomics
The soil-metal report refines the interpretation of corrected discovery counts by reporting that effect sizes have not been systematically examined across all 2,355 associations; a planned audit will inspect Spearman rho values and flag associations with rho < 0.05. 5 Planned partial-correlation tests, spatial-autocorrelation testing, and validation of unconditional versus conditional db-RDA R² are needed before the corrected associations can be interpreted as metal-specific or biologically substantial. 5 The report states that these validations require Spark access to the kescience_mgnify and kbase_ke_pangenome tables because no local CSV contains the Spearman rho values or model residuals. 5
Related pages
- Global Biogeography of Environmental Bacterial Metal Resistance — source summary reporting spatial and biome-corrected results. 1
- Community Metabolic Ecology via NMDC × Pangenome Integration — source summary reporting pathway–metabolite and ecosystem-comparison applications. 2
- PGP Gene Distribution Across Environments & Pangenomes — source summary reporting corrected PGP gene co-occurrence and environmental-enrichment analyses. 3
- Pan-Bacterial Anti-Phage Defense Arsenal — source summary reporting BH-FDR correction of phage-defense syndrome tests. 4
- Prophage Gene Modules and Terminase-Defined Lineages Across Bacterial Phylogeny and Environmental Gradients — source summary reporting corrected prophage module-environment and TerL-lineage tests. 6
- Carbon Source Utilization Predicts Ecology and Lifestyle in Pseudomonas — source summary reporting corrected carbon-pathway comparisons and ecological association tests. 7
- Soil Metal Concentrations Drive Functional Gene Shifts — source summary reporting corrected COG–metal associations and their validation limits. 5
- environmental-resistome — concept supported by biome-specific corrected prevalence and enrichment results. 1
- environment-embedding-geography — concept related to the spatial grid analysis and geographic interpretation of corrected hotspot results. 1
- phage-defense-syndromes-and-arms-race — concept related to corrected defense-system and prophage module enrichment tests. 4,6
- fishers-exact-test — statistical test used with this correction for hotspot, coldspot, and biome comparisons. 1