Lung model graph: curate full-text interaction candidates through the one-community research workflow #1461

CommunityMech:000432, FourSpecies_Aerobic_Lung_Microbiome_Model.yaml; PMID:42545019, doi:10.1128/msystems.00491-26, PMC13595950.

The record has no ecological_interactions. The openly available primary main body now supports measured consortium viability and pH phenotypes alongside explicitly proposed mechanisms. Species-specific vPCR supports day-10 persistence; total CFU, DNA-qPCR and protein-abundance fractions are distinct readouts. Pairwise cultures containing P. koreensis alkalinize, and the community recovers an alkaline pH after acid perturbation. The latter is functional, not member-resolved compositional resilience. Siderophore uptake, partner-resolved acid/pyruvate exchange, ammonia mediation and AI-2 coordination remain hypotheses, not direct flux or selective perturbation findings. Predicted antimicrobial pathways did not yield inhibition zones in the tested agar assay.

Acceptance: complete one-community Edison causal-graph dry run, obtain specific credit authorization, assess its cited evidence, and add only source-backed nodes/arrows with canonical four-species IDs. Added challenge organisms are not base-community members. Keep needs_research until new structure is actually curated; a gap alone is not completion. Preserve fixed DOI evidence from #1080. Full main-body text, tables and captions read; no figure images, supplements or raw data inspected. Dry run completed for one community, 7,671 query characters; no provider call or spend.

Deduplication: 778 all-state issue titles/bodies plus ignored/hidden-inclusive local reports, research, history and tests searched. All seven matched issue bodies read (#1358/#1356/#1173/#1171/#1170/#1169/#1080); relevant prior curation histories inspected. No exact duplicate within that scope; comments and PR reviews were not exhaustively searched. Source-based self-adversarial review, not independent approval.