Predicting allergenic hazards in novel insect protein sources using proteomics data

Reliable and sensitive analytical methods are required to assess the allergenicity hazard of novel alternative protein sources. Edible insects are increasingly introduced into the food market. In that context, allergenicity remains a key safety concern, particularly for individuals sensitive to crustaceans, mites, and mollusks. In the present study, we combined high-resolution mass spectrometry (HR-MS), proteomics, bioinformatics, and in-silico tools to map potential allergenic hazards associated with the ingestion of insect-derived proteins. We identified 2650 proteins from proteomics data from five insect species: black soldier fly (BSF) (Hermetia illucens), yellow mealworm (Tenebrio molitor), lesser mealworm (Alphitobius diaperinus), house cricket (Acheta domesticus), and morio worm (Zophobas morio).  Sequence-based screening against an in-house allergen database identified 454 putative allergenicity-associated protein hits. Among these, 40 highly abundant protein candidates were selected for subsequent in silico evaluation. AllerCatPro 2.0, which integrates amino acid sequence comparison and three-dimensional epitope similarity with curated allergen datasets, was used to rank proteins by predicted allergenic potential. For proteins exhibiting strong evidence of allergenicity but lacking matching structural epitopes, de novo three-dimensional models were generated using AlphaFold2. This approach enabled the identification and hazard ranking of candidate allergens based on abundance, sequence homology, and structural similarity. Overall, this study presents a proteomics-based workflow, combined with an in silico tool for the rapid identification and hazard ranking of candidate allergens, providing a practical framework to support allergenicity risk assessment of novel protein sources.