comp-005: Lactoferrin Shio-Koji Protease Stability¶

Question: Will human lactoferrin (P02788) survive the shio-koji protease environment with meaningful structural integrity retained?

Verdict: HIGH (full sequence) / MODERATE (mature protein aa 20–710)

The HIGH verdict is signal-peptide-contingent: all top-5 cleavage sites across all three koji proteases map to the fully disordered signal peptide (aa 1–19, pLDDT 35–54). If A. oryzae signal peptidase processes the heterologous signal sequence, the operative risk for the mature protein drops to MODERATE (ALP max 0.188). If signal peptide is retained, the N-terminus is fully exposed at HIGH risk. Signal peptide processing for heterologous secreted proteins in A. oryzae is common but not guaranteed.

Informs: validation-experiments.md §1.10 — lactoferrin arm remains a feasibility gate (unlike uricase arm, which comp-001 reframed as confirmation)

Interpretive wiki page: wiki/lactoferrin-protease-stability-computational.md

Companion experiment: comp-001 (uricase, LOW risk) — uses the same shared library

How to reproduce¶

cd experiments/comp-005-lactoferrin-shio-koji-protease-stability
python3 analyze.py

No external packages required (stdlib only: json, pathlib). Outputs land in outputs/.

The core algorithm lives in experiments/lib/protease_stability.py — shared with comp-001. This script is the orchestrator; the library exports functions only.

File index¶

comp-005-lactoferrin-shio-koji-protease-stability/
  analyze.py                     ← analysis script (run this)
  inputs/
    P02788.fasta                 ← human lactoferrin sequence (UniProt, 710 aa)
    alphafold_P02788_plddt.json  ← per-residue pLDDT scores (AF-P02788-F1-v6)
    protease_specificities.json  ← koji protease rules + shio-koji conditions
    provenance.md                ← sources, fetch dates, citations for every input
  outputs/                       ← generated by analyze.py; committed as artifacts
    cleavage_sites.json          ← machine-readable full results
    summary.md                   ← human-readable; cited in the wiki
  README.md                      ← this file

Key results¶

Scope	Verdict	Max risk score	Worst protease
Full sequence (incl. signal peptide)	HIGH	0.388	NPr
Mature protein only (aa 20–710)	MODERATE	0.188	ALP

Per-protease exposed sites:

Protease	Exposed sites (full)	Exposed sites (mature)	Effective activity
ALP (alkaline subtilisin)	21	3	18.8%
NPr (neutral metalloprotease)	14	0	38.8%
acid_protease (aspergillopepsin)	10	0	19.5%

All full-sequence exposed sites are in the signal peptide. ALP's 3 mature-protein exposed sites drive the MODERATE verdict for aa 20–710.

Disagreement protocol¶

If you reproduce the outputs and disagree with the methods or numbers, file a GitHub issue referencing this folder (comp-005-lactoferrin-shio-koji-protease-stability). Primary candidates for revision: signal peptide boundary (UniProt P02788 annotates aa 1–19; verify against A. oryzae signal peptidase cleavage rules), pLDDT thresholds for accessibility classification, or protease pH factors.