GOM oil graph: separate hydrocarbon depletion, gross removal and hypothesized cooperation #1475

CommunityMech:000030; GOM_Oil_Degrading_Consortium; doi:10.3389/fmars.2022.962071.

Publisher main-body Methods 2.4/2.7 and Results 3.3 distinguish the ~62% day-75 consortium gravimetric removal (vs ~25-39% individual cultures), for which abiotic control was NOT subtracted, from GC-MS biodegradation with an abiotic baseline. Each strain starts at 1e8 CFU/mL in both contexts, so consortium total starting CFUs exceed monoculture total. These comparisons do not isolate synergy from inoculum size. n-Alkane maximum ~60% at day30 falls to ~45% at day75; PAH endpoint ~16% and other-hydrocarbon mass-balance estimates are different metrics. Stable relative abundance after day7 and MetaMIS interaction predictions are not measured partner-specific flux or proof of niche partitioning. Degradation of some methylated phenanthrenes is not generic phenanthrene degradation.

Acceptance: preserve three bounded measured-process/outcome nodes; withdraw unsupported NICHE_PARTITIONING classifications and both class-depletion-to-enhanced-bulk-removal causal arrows, which do not selectively explain the comparative gross-removal advantage. Bound representative compounds and exact four canonical participants; distinguish inferential roles from measurement. Add an explicit inoculum/abiotic/mediation knowledge gap and retain potential modeled network extension as uncurated, not proven. Existing cache contains the abstract plus one Methods sentence despite its FULL TEXT marker. Primary publisher body was separately read; no cache expansion just to satisfy snippet checks. Guarded history, tests, generated page and review outputs required. No new causal structure or provider spend required.

Source-based self-adversarial review, not independent approval. All three whole YAML records were read. Main text, Methods and embedded captions were read; no figure images, separate supplementary files, raw data or analysis code audited. Dedup searched 790 all-state issue titles/bodies and ignored/hidden-inclusive reports, research, history, tests, scripts and workflows. Five matched issue bodies (#624/#543/#486/#326/#17) and the UFMP/Garlic prior history artifacts were read; none duplicates these source-specific causal-scope repairs. Existing validation and render logs also match record names but do not constitute completed source reviews. Not every issue comment or PR review was inspected.