A target-priority score ranks taxa by how strongly they are implicated in disease, but it says nothing about what the community loses when that taxon is removed. In the IBD phage-targeting work the dominant such cost axis is explicit: the analysis identifies a bile-acid 7α-dehydroxylation network — the conversion of primary bile acids into secondary, unconjugated forms — comprising flavonifractor plautii, eggerthella lenta and enterocloster bolteae as its principal per-target ecological-cost axis in paired hmp2 samples 1.
Per-target annotation axes a headline score does not carry
The cross-project digest argues that pathobiont-targeting interventions need three axes of per-target annotation alongside the headline Tier-A priority score: the iron/AIEC (adherent-invasive E. coli) versus other Crohn’s-disease specialization mechanism, the bile-acid coupling cost, and whether an association is mediated by species abundance or by strain content 2. This is a design caveat rather than a measurement: it states what a ranking must be annotated with, not a result about any one taxon 2.
The first axis is a question of targeting granularity: distinguishing an iron/AIEC-mediated specialization mechanism from other Crohn’s-disease mechanisms informs whether an intervention should target an entire species broadly or specifically its AIEC subset 2. Framed that way, the mechanism question developed in iron acquisition and pathobiont expansion becomes an operational choice about target resolution, and it refines the translation problem framed in phage-therapy-evidence-translation by making target breadth, not just target identity, the decision 2.
The bile-acid 7α-dehydroxylation network
The second axis is instantiated concretely. F. plautii, E. lenta and E. bolteae are reported as active bile-acid-7alpha-dehydroxylation taxa, while mediterraneibacter gnavus and escherichia-coli are not; because of this, depleting active 7α-dehydroxylators is expected to shift the bile-acid pool toward inflammatory primary tauro-conjugated forms 2. That last step is an inference from substrate–product structure rather than a measured intervention outcome, so it should be read as the hypothesis that removing an active dehydroxylator reverses the direction of the pool it maintains — a prediction that connects target choice to the inflammatory metabolite classes catalogued in metabolite-class-signatures-of-gut-inflammation 2.
The network assignment rests on correlational evidence. In 468 paired CSM* HMP2 samples, F. plautii, E. lenta and E. bolteae show the predicted substrate–product signature: negative spearman correlation with primary tauro-conjugated bile acids (substrates: cholate at ρ=−0.26 for F. plautii) and positive correlation with secondary unconjugated bile acids (products: lithocholate at ρ=+0.18 for E. bolteae, +0.15 for F. plautii; deoxycholate +0.17 for E. bolteae) 1. These are observational sample-level correlations within one cohort rather than enzymatic assays, which places the membership call at the level of a detectable substrate–product signature, as the project itself describes it 1.
The negative side of the comparison carries its own information: M. gnavus and E. coli show the opposite correlation pattern, confirming they are not in the 7α-dehydroxylation network 1. This supports the digest’s split of the same five taxa into active and inactive dehydroxylators 2, and it means the bile-acid coupling cost is an axis that discriminates among candidate targets rather than a blanket penalty on pathobiont depletion 2. It does not license the stronger reading that depleting an inactive taxon leaves bile-acid metabolism untouched: exclusion from this network means the direct substrate–product coupling was not detected, and no source here tests indirect effects of removing M. gnavus or E. coli on the pool 1.
Within the project, this axis displaced an earlier one. The prior “metabolic-coupling cost” annotation was based on species-level co-occurrence, and the paired sample-level test failed to identify a metabolic intermediate that supports cross-feeding — lactate has opposite signs between A. caccae and the pathobionts, with only 7 strict triangles total — so the bile-acid network was adopted in its place as the more mechanistically grounded cost annotation 1. The replacement is a change of evidence class, from co-occurrence to a sample-level substrate–product signature, not a contradiction between two findings 1.
Open Directions
- Test the depletion prediction directly rather than by correlation: in a gnotobiotic or ex vivo community, remove an active dehydroxylator (F. plautii, E. lenta or E. bolteae) and measure whether the bile-acid pool shifts toward primary tauro-conjugated forms, closing the gap between a substrate–product correlation and a causal ecological cost 1.
- Rank the three annotation axes against each other by scoring the same candidate target list on iron/AIEC mechanism, bile-acid coupling and species-abundance-versus-strain-content mediation, and checking whether they reorder targets concordantly or pull in different directions — the digest requires all three but does not say which binds 2.
- Resolve the AIEC-subset question at strain resolution: stratify the E. coli signal into AIEC and non-AIEC strain content within the same cohort to establish whether species-level depletion is warranted, which is the choice the mechanism axis is meant to decide 2.
- Measure whether depleting a taxon outside the network (M. gnavus or E. coli) perturbs the bile-acid pool indirectly, since the opposite-sign correlations establish only that these taxa lack the direct substrate–product signature 1.
- Replicate the reported 468 paired CSM* correlations in an independent cohort; that specific quantitative support comes from a single paired metagenome–metabolome comparison, even though the project frames its bile-acid result as a multi-line convergence narrative 1.