Treffer: Allergen Chip Challenge: A nationwide open database supporting allergy prediction algorithms.

Background: Allergen chip (AC) technologies are a powerful tool for simultaneous analysis of hundreds of allergens, generating a comprehensive sensitization landscape for precision medicine in allergy. These considerable data require extensive knowledge for translation into clinically relevant conclusion.
Objective: To harness machine learning for AC interpretation in daily practice, we set out to establish a nationwide open database of AC, demographic, and clinical information and to submit it to an international crowdsourced machine learning competition to generate a predictive allergy classification algorithm.
Methods: The project consortium defined 20 clinical variables and 5 demographic factors for retrospective collection in conjunction with AC IgE data (2014-23) from 11 French university hospitals. The dataset was processed to tag confirmed allergy, grade of severity, and culprit allergen identification associated with AC data and submitted to the data challenge.
Results: Data were collected for 4271 patients, yielding a dataset with over 700,000 specific IgE data points. Sensitization was present in 3579 patients (84%). Allergy was confirmed in 2236 patients (53%) and excluded in 1076 patients, with the remaining 959 being missing outcome data (allergy diagnosis labels). The competition attracted 292 data scientists who submitted 3135 algorithms. The highest F scores ranged from 0.780 to 0.786. The database was subsequently made available as open source.
Conclusions: We present a nationwide open allergy database designed to enable the development of predictive algorithms. This scalable framework, integrating clinical data with machine learning techniques, paves the way for data-driven AC use and interpretation by allergists.
(Copyright © 2025 The Authors. Published by Elsevier Inc. All rights reserved.)

Disclosure statement The Allergen Chip Challenge project was funded by the French Bank for Public Investment, the French National Centre for Scientific Research, and the French National Program for Research in Artificial Intelligence (grant 2023). The project received institutional support from the Health Data Hub, a public interest group supported by the French Ministry of Health to facilitate the creation and the use of the database by research and development teams. Open-source implementation of database: The database (doi.org/10.60597/j5fe-g420) presented in this report is freely available and can be downloaded from www.data.gouv.fr/fr/datasets/allergen-chip-challenge. The open-license Etalab was chosen to facilitate and encourage reuse of public data made available free of charge. At present, users are at liberty to extract and analyze data without restriction, but it should be noted that no open input is permitted. This means that users are authorized to access and extract data without restriction; however, modification of the original dataset is currently prohibited without the authorization of the database governance committee. Disclosure of potential conflict of interest: J. Vitte reports, outside the current study, speaker and consultancy fees and travel support from Novartis, Sanofi, Stallergenes-Greer, Thermo Fisher Scientific, and Zambon. J. Goret reports, outside the current study, speaker and travel support from ALK, Stallergenes-Greer, Thermo Fisher Scientific, and Menarini. J. Corriger reports, outside the current study, speaker and travel support from ALK, Stallergenes-Greer, Thermo Fisher Scientific, and Menarini. The rest of the authors declare that they have no relevant conflicts of interest.