{"schema_version":1,"research_id":"eoa_inverse_innovation_exp05_external_evaluation_20260803","source_assessment_id":"invariant_mode_decomposition_design__biology_ecology:P4:v0","cell_id":"invariant_mode_decomposition_design__biology_ecology","search_queries":["conservation breeding mean kinship optimal contribution genetic management founder genome equivalents zoo population planning PMx official","AZA population management center breeding transfer plan mean kinship official","conservation breeding optimal contribution selection captive populations founder representation research","spectral analysis population genetics founder contributions breeding transition matrix eigenvalue conservation breeding","site:species360.org PMx population management software mean kinship founder genome equivalents manual","PMx software conservation breeding population management mean kinship official manual PDF","site:aza.org animal welfare breeding transfer plan institutional authority policy responsible population management PDF","AZA breeding transfer recommendations implementation fulfillment rate study 20% 57% Zoo Biology 2023","\"Reasons for unfulfilled breeding and transfer recommendations\" DOI","\"PMx: software package for demographic and genetic analysis\" DOI full text","\"Optimum contribution selection for conserved populations with historic migration\" DOI","(\"eigenvalue\" OR spectral OR \"invariant mode\") \"founder contribution\" breeding conservation","(\"transition matrix\" AND founder) conservation breeding plan genetic contribution","\"modal\" \"mean kinship\" zoo breeding","\"spectral\" conservation breeding \"founder\""],"sources":[{"source_id":"S1","title":"Species Survival Plan Programs","publisher":"Association of Zoos and Aquariums","url":"https://www.aza.org/species-survival-plan-programs","source_class":"OFFICIAL_ORGANIZATION_DATA","publication_date":"n.d.; current page accessed 2026-08-03","accessed_at":"2026-08-03","claims_supported":["AZA reports nearly 300 Species Survival Plan programs.","SSP programs develop studbooks and Breeding and Transfer Plans to manage genetically diverse, demographically varied populations.","Identifiable adopters include SSP program leaders, the AZA Population Management Center, and participating institutions."]},{"source_id":"S2","title":"PMx: software package for demographic and genetic analysis and management of pedigreed populations","publisher":"Methods in Ecology and Evolution, British Ecological Society/Wiley","url":"https://besjournals.onlinelibrary.wiley.com/doi/10.1111/j.2041-210X.2011.00148.x","source_class":"PRIMARY_RESEARCH","publication_date":"2012","accessed_at":"2026-08-03","claims_supported":["PMx already analyzes pedigree, demographic, kinship, founder-contribution, founder-genome-equivalent, fertility, location, and reproductive-status data.","PMx supports genetic-diversity analysis, pairing recommendations, founder-representation summaries, and simulated allele transmission.","The candidate would extend an established technical workflow rather than create conservation-population analysis from scratch."]},{"source_id":"S3","title":"Reasons for unfulfilled breeding and transfer recommendations in zoos and aquariums","publisher":"Zoo Biology/Wiley","url":"https://onlinelibrary.wiley.com/doi/abs/10.1002/zoo.21664","source_class":"PRIMARY_RESEARCH","publication_date":"2021-12-06","accessed_at":"2026-08-03","claims_supported":["Action-requiring recommendations historically had much lower fulfillment than Hold or Do Not Breed recommendations.","Unfulfilled breeding recommendations commonly reflected biological and behavioral causes; transfers also faced institutional, communication, shipment, permit, financial, and data-management constraints.","A deterministic expected transition operator must represent uncertainty and cannot be treated as a promise that recommendations will be realized."]},{"source_id":"S4","title":"The influence of captive breeding management on founder representation and inbreeding in the ‘Alalā, the Hawaiian crow","publisher":"Conservation Genetics/Springer Nature","url":"https://asu.elsevierpure.com/en/publications/the-influence-of-captive-breeding-management-on-founder-represent-2/","source_class":"PRIMARY_RESEARCH","publication_date":"2016-04-01","accessed_at":"2026-08-03","claims_supported":["A real conservation population retained severe founder imbalance: 45% of 2014 ancestry came from one founder pair, and 76% of inbreeding was attributed to that pair.","Managed breeding had evened some founder contributions and maintained relatively low overall inbreeding, while substantial lineage concentration persisted.","Founder representation can therefore matter materially even under active breeding management."]},{"source_id":"S5","title":"Optimum contribution selection for conserved populations with historic migration","publisher":"Genetics Selection Evolution/BMC","url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC3807754/","source_class":"PRIMARY_RESEARCH","publication_date":"2012","accessed_at":"2026-08-03","claims_supported":["Optimal-contribution selection already calculates breeding contributions under gene-diversity and founder-origin objectives.","The choice of objective function can materially change founder contributions and conservation outcomes.","This is a close comparator to any proposed founder-contribution plan gate."]},{"source_id":"S6","title":"A unifying concept of animal breeding programmes","publisher":"Journal of Animal Breeding and Genetics/Wiley","url":"https://onlinelibrary.wiley.com/doi/full/10.1111/jbg.12534","source_class":"PRIMARY_RESEARCH","publication_date":"2021","accessed_at":"2026-08-03","claims_supported":["Animal-breeding gene-flow methods already represent sex-, age-, and selection-defined cohorts with transition matrices through reproduction and aging.","Matrix-based deterministic breeding-program simulations are established prior art.","The use of a complete transition operator is therefore not independently novel; the remaining contrast concerns founder-specific spectral gating, validation, and governance."]},{"source_id":"S7","title":"EAZA Population Management Manual: Standards, procedures and guidelines for population management within EAZA","publisher":"European Association of Zoos and Aquaria","url":"https://media.eaza.net/uploads/EAZA_Population_Management_Manual_V6_2_67db7a6627.pdf","source_class":"OFFICIAL_GUIDANCE","publication_date":"2025-01","accessed_at":"2026-08-03","claims_supported":["EAZA requires recurring demographic and genetic analyses and identifies PMx as an available analysis tool.","Coordinators must compromise among genetic, age, health, social, reproductive, transport-cost, exhibit, and other practical considerations.","Studbook-data access is role-controlled and external access may require approval and a nondisclosure agreement.","The manual warns that deterministic projections omit random events and states that institutional decisions and local law remain controlling."]},{"source_id":"S8","title":"AZA Policy on Responsible Population Management","publisher":"Association of Zoos and Aquariums","url":"https://assets.speakcdn.com/assets/2332/aza_policy_on_responsible_population_management_final.pdf","source_class":"OFFICIAL_GUIDANCE","publication_date":"2016-01-12","accessed_at":"2026-08-03","claims_supported":["Animal health and welfare must remain priorities, and applicable laws, ownership, chain-of-custody, and records requirements apply.","An institution's director or chief executive has final authority for acquisitions, transfers, and euthanasia.","Transfers and other population actions require institutional documentation; an advisory model cannot itself authorize animal action."]}],"problem_evidence":{"support":"STRONG","rationale":"The ‘Alalā case directly demonstrates consequential, persistent founder concentration within an actively managed conservation population, while AZA and EAZA sources confirm that maintaining genetic diversity and demographic stability is a recurring operational objective. The evidence establishes the general problem, but not the prevalence of hidden growing eigenmodes across programs.","source_ids":["S1","S4","S7"]},"stakeholder_evidence":{"support":"MODERATE","rationale":"AZA SSPs and EAZA EEPs are identifiable adopters with formal responsibility for breeding and transfer recommendations and an expressed need for recurring genetic analysis. However, no source expresses demand specifically for an eigenmode gate, and institutional uptake or funding commitment has not been demonstrated.","source_ids":["S1","S7","S8"]},"prior_art":{"proximity":"ADJACENT_PRIOR_ART","closest_analogues":[{"name":"PMx population-management analysis","similarity":"Uses the same pedigree, founder-contribution, kinship, demographic, fertility, location, and recommendation data and already reports founder representation and simulates inheritance.","remaining_difference":"The reviewed PMx description does not condition plan authorization on eigenmode gain, spectral separation, conditioning, held-out modal reconstruction, or drift.","source_ids":["S2","S7"]},{"name":"Optimum contribution selection","similarity":"Optimizes candidate reproductive contributions under aggregate coancestry, gene-diversity, and founder-origin objectives.","remaining_difference":"It optimizes an objective over contributions rather than testing stability and sensitivity of reproducible founder-by-class modes of a repeated complete-plan operator.","source_ids":["S5"]},{"name":"Gene-flow transition-matrix breeding models","similarity":"Already model reproduction and aging as cohort transition matrices and calculate expected genetic flow through complex breeding programs.","remaining_difference":"The located practice concerns expected genetic progress and breeding-program description, not a conservation authorization gate for founder-concentration modes with residual, conditioning, drift, and welfare checks.","source_ids":["S6"]},{"name":"AZA/EAZA breeding and transfer planning","similarity":"Already combines genetic analysis with demographic, welfare, health, social, institutional, and logistics constraints to issue plan-level recommendations.","remaining_difference":"The official workflows reviewed do not describe a spectral founder-mode comparator or a preregistered modal pass/fail criterion.","source_ids":["S1","S7","S8"]}],"distinctive_claim_remaining":"For one bounded conservation population and planning horizon, after identical welfare, veterinary, demographic, legal, and institutional constraints are imposed, a preregistered founder-by-class transition-operator gate will identify a reproducible, adequately conditioned concentration mode and select a feasible shadow plan that reduces that mode's held-out realized coordinate or prediction error relative to PMx/minimum-mean-kinship, optimal-contribution, and stochastic endpoint comparators; the claim fails if mode identity or ranking is unstable, no constrained plan distinction remains, or comparator performance is not improved.","confidence":"MODERATE"},"implementation_evidence":{"support":"MODERATE","rationale":"Existing PMx/studbook infrastructure, established cohort transition matrices, recurring planning workflows, and explicit institutional authority make a retrospective shadow implementation technically and procedurally plausible. Feasibility remains unverified for real SSP data: stochastic reproduction, pedigree error, incomplete founder relationships, non-normal transient growth, small spectral gaps, protected data access, and inconsistent recommendation fulfillment could invalidate the operator or its interpretation. Safety is manageable only because the first test is non-operative and institutional authority remains unchanged.","source_ids":["S2","S3","S6","S7","S8"]},"scores":{"meaningful_impact":{"score":4,"rationale":"Persistent founder concentration can account for a large share of ancestry and inbreeding in an endangered managed population, so earlier detection could materially protect genetic-management objectives.","source_ids":["S4"]},"stakeholder_pull":{"score":3,"rationale":"Formal programs and decision owners exist and require genetic analyses, but mode-specific demand, budget, and partner commitment are absent.","source_ids":["S1","S7"]},"incremental_advantage":{"score":3,"rationale":"The gate could expose multi-cycle directional behavior that static founder summaries or aggregate objectives do not, but superiority over existing optimization and simulation is wholly untested.","source_ids":["S2","S5","S6"]},"distinctiveness_plausibility":{"score":3,"rationale":"The reviewed sources establish all major building blocks—founder analysis, contribution optimization, transition matrices, and plan governance—but not their combination as a validated spectral authorization gate.","source_ids":["S2","S5","S6","S7"]},"technical_implementability":{"score":3,"rationale":"Matrix construction and decomposition are computationally routine, but a biologically valid, stable operator from sparse and stochastic breeding records is not yet demonstrated.","source_ids":["S2","S3","S6","S7"]},"adoption_authority_feasibility":{"score":3,"rationale":"SSP/EEP planners are credible advisory adopters and institutional executives retain final action authority; multi-institutional consent, data access, and workflow integration remain substantial dependencies.","source_ids":["S1","S7","S8"]},"evidence_readiness":{"score":2,"rationale":"The test can be specified now, but it requires protected historical studbook records, comparator implementations, partner expertise, and completed-cycle outcomes not available in the reviewed public evidence.","source_ids":["S2","S3","S7"]},"safety_net_benefit":{"score":4,"rationale":"The proposal targets genetically vulnerable managed populations and its shadow-only first step adds a reversible warning layer without displacing welfare, veterinary, legal, or institutional controls.","source_ids":["S4","S7","S8"]},"scalability":{"score":3,"rationale":"Reusable software could transfer across pedigreed programs, but every species requires state definitions, fertility assumptions, constraint encoding, validation thresholds, and governance approval.","source_ids":["S1","S2","S3","S7"]}},"score_confidence":"MODERATE","costs":{"first_evidence":{"band_2026_usd":"50K_TO_250K","scope":"One-program retrospective replay covering data-governance setup, historical cutoff, operator construction, three comparator implementations, uncertainty analysis, held-out evaluation, and a non-operative shadow report.","confidence":"LOW","assumptions":["Existing studbook and completed-cycle records can be exported without new animal sampling.","Approximately 0.5-1.5 combined FTE-years from a population geneticist, quantitative modeler, data steward, and program coordinator.","No operational animal action, transfer, or genomic assay is included."],"source_ids":["S2","S3","S7"]},"initial_deployment_startup":{"band_2026_usd":"250K_TO_1M","scope":"Productionize the method for 2-3 partner programs with secure ingestion, audit trails, uncertainty and non-normality diagnostics, comparator pipelines, documentation, training, and independent validation.","confidence":"LOW","assumptions":["Existing PMx/ZIMS exports remain the primary data interface.","Species-specific model configuration is required but no new enterprise studbook platform is built.","Institutional legal and welfare review uses existing committees and policies."],"source_ids":["S2","S7","S8"]},"operational_launch":{"band_2026_usd":"1M_TO_5M","scope":"Regional launch across approximately 10-25 programs, including integration, program-specific validation, coordinator training, governance review, support, and an external evaluation reserve.","confidence":"LOW","assumptions":["Launch remains advisory and does not automate authorization.","Each program needs a separate model contract, data agreement, and acceptance test.","The range excludes animal transport, breeding operations, facility construction, and genomic-data generation."],"source_ids":["S1","S3","S7","S8"]},"annual_recurring":{"band_2026_usd":"250K_TO_1M","scope":"Maintain a 10-25-program advisory service with scheduled recomputation, data-quality review, drift and residual monitoring, incident review, software maintenance, and annual governance reporting.","confidence":"LOW","assumptions":["One small central technical team is shared across programs.","Program coordinators and institutional reviewers contribute part-time through existing roles.","Major redevelopment, new genomic assays, and animal-care costs are excluded."],"source_ids":["S1","S3","S7"]}},"verified_pipeline_gates":{"externally_supported_problem":{"status":"YES","reason":"A primary conservation case documents persistent founder-pair concentration and associated inbreeding despite active management.","source_ids":["S4"]},"externally_credible_adopter_or_authorizer":{"status":"YES","reason":"AZA SSP and EAZA EEP structures provide identifiable advisory adopters, while official policy identifies institutional final authority.","source_ids":["S1","S7","S8"]},"distinct_testable_incremental_claim":{"status":"YES","reason":"The remaining claim is bounded to reproducibility, held-out prediction or targeted-coordinate reduction, and plan distinction against named comparators under matched protected constraints.","source_ids":["S2","S5","S6","S7"]},"bounded_next_evidence_step":{"status":"YES","reason":"A single-program, historical-cutoff, non-operative replay with preregistered thresholds and comparators can answer the first evidence question without animal action.","source_ids":["S2","S3","S7"]},"no_unresolved_safety_or_authority_stop":{"status":"YES","reason":"For the shadow test, outputs have no operational force; welfare, legal, veterinary, and institutional authority remain independently controlling. Operational use would require renewed local approval.","source_ids":["S7","S8"]},"credible_cost_scope_and_range":{"status":"UNCERTAIN","reason":"Work packages and exclusions can be bounded from the documented workflow, but no public partner quote, labor estimate, integration assessment, or data-access cost was found.","source_ids":["S2","S7"]}},"next_evidence_step":"Recruit one SSP or EEP partner and preregister a retrospective, non-operative replay. Freeze all records before a historical cutoff; define the founder-by-reproductive-class state, reference distribution, candidate-plan constraints, fertility/retention uncertainty ranges, operator construction, mode-matching method, eigenvector condition limit, non-normal transient-growth diagnostic, spectral-gap threshold, and residual budget. Without refitting, score at least two subsequent completed cycles and bootstrap pedigree and fertility uncertainty. Compare four identically constrained planners: current PMx/minimum-mean-kinship practice, optimal-contribution selection, stochastic Monte Carlo endpoint ranking, and the modal gate. Primary outcomes are mode identity stability, held-out founder-coordinate error, residual structure, whether plan ranking changes, and whether the targeted realized coordinate improves. Falsify the proposal if no action-relevant mode is reproducible, the modal plan is not distinct, it fails to outperform the best comparator, its advantage disappears under uncertainty or protected constraints, transient growth contradicts the eigenvalue classification, or required data provenance or authorization cannot be secured. No breeding, transfer, contraception, retention, or other animal action follows from the test.","blocking_evidence":["No public study establishes how often complete conservation breeding plans contain hidden persistent or growing founder-contribution modes beyond standard founder and kinship summaries.","No reviewed source validates an eigenmode authorization gate on completed conservation-breeding cycles.","No SSP/EEP partner commitment, data-use approval, or access to protected studbook and outcome records has been demonstrated.","The proposed founder-by-class transition operator, uncertainty treatment, mode-matching rule, conditioning threshold, spectral-gap threshold, and residual budget have not been validated on real program data.","Comparative performance against PMx/minimum mean kinship, optimal-contribution selection, and stochastic endpoint simulation is unknown.","Planner comprehension, trust, and resistance to automation bias have not been tested.","Jurisdiction- and institution-specific legal, contractual, ownership, privacy, and welfare review remains undone.","Cost bands lack partner estimates, procurement quotes, and measured integration effort."],"research_disposition":"PARTNERED_RESEARCH_PROGRAM","world_novelty_boundary":"World novelty, patentability, freedom to operate, market size, and realized impact were not measured. The bounded search found established founder-representation analysis, PMx recommendations, optimal-contribution selection, breeding-program transition matrices, and formal plan governance, but did not find a conservation-breeding practice that gates a complete plan on founder-specific modal gain, conditioning, spectral separation, held-out reconstruction, and drift. This is only a contrast within the eight reviewed direct sources and recorded queries, not a world-novelty conclusion.","arm":"COMPLETE_PROPOSAL_PORTFOLIO","candidate_version":0,"controller_recommendation":{"action":"STOP_EMPIRICAL_RESEARCH_NEEDED","repairable":false,"material_progress_observed":true,"progress_targets":["Secure one SSP or EEP partner, data-use authorization, and a documented institutional authority map.","Obtain de-identified pre-cutoff studbook, eligibility, plan, transfer, retention, and completed-cycle outcome data for at least two held-out cycles.","Preregister the operator, uncertainty model, non-normality diagnostic, mode-matching rule, thresholds, comparators, outcomes, and falsifiers.","Demonstrate stable modes and acceptable held-out residuals before interpreting any plan ranking.","Show incremental performance against current PMx/minimum-mean-kinship, optimal-contribution, and stochastic-simulation comparators under identical welfare and demographic constraints.","Obtain partner-based staffing, integration, security, and recurring-support estimates to validate the cost bands.","Conduct planner usability and automation-bias review before any operational consideration."],"reason":"Bounded web research establishes the problem, credible authorities, adjacent prior art, and a testable residual claim, but the decisive evidence requires protected program records, partner participation, species-specific modeling, and retrospective empirical validation. Those questions cannot be repaired by further public-web research alone."},"proposal_index":4}