Lactoferrin Inter-Lobe Linker Candidate Design (comp-034)¶

The engineering question is whether an exact inter-lobe connector can be altered without disrupting lactoferrin's bilobal fold, and whether such a change improves retained function through shio-koji processing.

COMP-034 cannot answer that question. It reused an unverified legacy protease-preference table as a cleavage axis, then combined it with uncalibrated sequence, ProteinMPNN, and Rosetta scores. The resulting cleavage values, GREEN/STRICT tiers, concordance claims, candidate ordering, and wet-lab priorities are invalid. The COMP-034 tombstone is non-runnable.

Research conjecture — an observed linker-associated failure may be engineerable

Grounded premises: UniProt P02788 and PDB 1B0L define the short connector between the two lactoferrin lobes (In Vitro structural record; PMID 10089347). Section §1.10 treats fragment formation and retained iron binding as unmeasured empirical gates.

Novel leap: If a reproducible WT fragment maps to this connector, a sequence change might reduce that failure while preserving lobe geometry and lactoferrin function. No direct study has tested that failure-specific redesign in shio-koji.

Why it matters: A successful redesign could preserve the payload and delivery route without deleting a structurally required connector.

Discriminating observation: First establish a reproducible WT fragment and map its termini. Only then run a new gated design against verified specificity and structural constraints, followed by a matched diversity panel measuring intact abundance, fragment pattern, iron binding, and thermal stability.

No variant is currently selected. A negative WT fragmentation result redirects effort away from linker engineering without rejecting lactoferrin or the broader delivery portfolio.