About AllergyPred

AllergyPred is a freely available webserver designed to predict potential food allergens, encompassing both protein and chemical-based allergens.

AllergyPred is an in silico tool designed for the prediction of food allergenicity in proteins and chemical compounds. It leverages advanced computational models, including machine learning, to assess the likelihood of a substance causing allergic reactions. The tool is particularly useful in the context of food safety, pharmaceutical product development, and regulatory assessments, as it minimizes reliance on traditional animal testing and supports quicker evaluations. Results are presented in a user-friendly table format, downloadable in several file formats, facilitating seamless integration into research or project reports.

Allergy Models

machine learning models image
Computational Allergy Prediction

AllergyPred combines machine learning, bioinformatics, cheminformatics, statistics and immunology to assess potential allergenicity of proteins and chemical compounds such as food proteins, drugs, chemicals, or environmental agents.
AllergyPred uses labeled data on allergens and non-allergens to train respective classification models.
Using feature selection techniques, relevant features are extracted - such as amino acid composition, physicochemical properties, or epitope profiles. Algorithms includes support vector machine, deep neural network. It also provide a likelihood score for allergenicity.

Prediction Pipeline

prediction pipeline image
Citations

Banerjee Priyanka , Dehnbostel Frederic O. , Preissner Robert. Prediction Is a Balancing Act: Importance of Sampling Methods to Balance Sensitivity and Specificity of Predictive Models Based on Imbalanced Chemical Data Sets , Frontiers in Chemistry, 6, 2018, PMID: 30271769, https://doi.org/10.3389/fchem.2018.00362.

Worm, M., Alexiou, A., Höfer, V., Birkner, T., Jeanrenaud, A. C. S. N., Fauchère, F., Pazur, K., Steinert, C., Arnau-Soler, A., Banerjee, P., et al. An interdisciplinary approach to characterize peanut-allergic patients-First data from the FOOD@ consortium, Clinical and translational allergy, 12(10), 2022, e12197, PMID: 36225266, https://doi.org/10.1002/clt2.12197.

Emanuel Kemmler, Julian Braun, Florent Fauchère, Sabine Dölle-Bierke, Kirsten Beyer, Robert Preissner, Margitta Worm, Priyanka Banerjee. Data-driven analysis of chemicals, proteins and pathways associated with peanut allergy: from molecular networking to biological interpretation, Food Science and Human Wellness, Volume 13, Issue 3, 2024, Pages 1322-1335, ISSN 2213-4530, https://doi.org/10.26599/FSHW.2022.9250111.