Emiliania graph: separate observed host effects from unproven endogenous IAA mediation #1453

Emiliania_Phaeobacter_Dynamic_Interaction.yaml, PMID:27855786.

Bulk IAA was not detected in coculture; exogenous dose responses are not measured phycosphere exposure. Putative IAA-pathway mutants retained IAA and bleaching, so they are not IAA-null necessity controls. Isotope incorporation demonstrates added tryptophan conversion in bacterial culture, not the complete endogenous algal-to-bacterial-to-algal flux. Remove the generic nutrient-growth-to-specific-tryptophan workflow arrow; retain a HYPOTHESIZED tryptophan/IAA-to-host-response arrow with the missing mediation controls explicit. Keep nutrient support, tryptophan-associated attachment/IAA production, and observed algal promotion/late killing as three existing nodes. Remove unsupported categorical mutualism/predation and the tryptophan-biosynthesis annotation from precursor uptake/conversion. Do not add E. coli reporter or CCMP372 comparator to the two-member community. Roseobacticide activity in CCMP372 does not establish killing of CCMP3266. Preserve the nonpermissive-medium context (#583/#529).

Review scope: whole records and cached primary main bodies, tables and captions read. No figure images, separate supplements or raw data reviewed. Selected Emiliania published peer-review discussion was also read, not the entire exchange. Deduplication searched 772 all-state issue titles/bodies and ignored/hidden-inclusive local content; #583/#529 and #1091/#1094 bodies reviewed. No exact duplicate within that scope; not every issue comment or PR reviewed. This is source-based self-adversarial review, not independent approval.