Contributions
Metabolic Capability vs Metabolic Dependency compared genomic metabolic capability with measured metabolic dependency across 1,695 pathway-organism pairs from 48 organisms 1. It used GapMind pathway completeness predictions, Fitness Browser gene-fitness data, and SEED subsystem annotations as a pathway-membership proxy 1. The project found 267 (15.8%) latent pairs, meaning complete pathways with no detectable fitness importance under the tested conditions, alongside 547 (32.3%) intermediate and 881 (51.9%) active pairs 1. Pathway category strongly predicted dependency class (χ²=163.6, df=4, p=2.5×10⁻³⁴) 1. Carbon source utilization pathways were 24.3% latent and amino acid biosynthesis pathways were 6.5% latent, making carbon pathways 3.7× more likely to be latent 1. The project framed a prediction that organisms in nutrient-rich environments should disproportionately lose carbon-pathway genes, but it did not test that prediction 1. The latent fraction ranged from 0–31.6% across organisms, with a mean of 15.0% 1. A sensitivity analysis over 16 threshold combinations gave latent fractions from 4.7% to 21.1% (SD = 5.9 percentage points), and the qualitative conclusion held across that range 1.
The same project tested whether latent capabilities track conservation and pangenome openness 1. Pathway-level conservation did not distinguish latent from active pathways: mean conservation was 0.869 for latent pathways and 0.829 for active pathways (Mann–Whitney U p = 0.94 for active > latent, rank-biserial r = 0.052), so H2a was not supported 1. After aggregating 41 organisms to 22 species clades, the project found that latent capability rate correlated positively with pangenome openness (Spearman ρ = 0.69, p = 0.0004, n = 22 clades), supporting H2b 1. The report read these results as mixed support for the Black Queen framework, the idea that costly functions are lost when neighbours supply them, and as consistent with a signal operating at the level of genome dynamics and community context 1. Across 7,334 species, it reported Klebsiella pneumoniae as the most open well-sampled pathogen (99.05% open) and Mycobacterium tuberculosis as the most closed (97.4% open) 1.
The project also clustered within-species pathway profiles in 10 target species using hierarchical clustering and principal component analysis 1. All 10 showed metabolic clustering with silhouette scores (a measure of how well-separated clusters are) greater than 0.2, ranging from 0.349 for PALSA-747 sp. to 0.894 for Salmonella enterica (11,396 genomes, k=6, where k is the number of clusters) 1. Clusters were significantly associated with isolation environment in Salmonella enterica (χ²=1570.2, df=25, p<0.0001) and Phenylobacterium sp. (χ²=12.2, df=1, p=0.0005) 1. Associations were not significant in the marine organisms Stutzerimonas, Alteromonas, Pelagibacter, and Prochlorococcus, which the project suggested may reflect insufficient or coarse isolation-source metadata, depth- or nutrient-zone structure, or pathway annotations lacking resolution for relevant gene-content differences 1. The project stated that these associations are observational and may be confounded by phylogeny, because no explicit phylogenetic correction was performed 1.
The project reported several limitations 1. SEED-proxy mapping can introduce false-positive genes that dilute the fitness signal 1. Laboratory-biased fitness conditions may inflate the latent fraction 1. Only 48 fitness-tested organisms were available among 293,000 genomes, which limits how far the results generalize 1. An organism-to-clade linkage through gtdb_metadata taxids returned zero matches because the relevant column contained boolean strings rather than numeric taxids, so downstream analyses used organism-level fitness aggregates without an explicit clade-level linkage, which may reduce the precision of the H2a comparison 1. The deoxyribonate and myoinositol pathways were excluded from all analyses because they had no matching SEED role descriptions 1. Alanine was excluded from classification for every organism because of insufficient SEED coverage 1.