Databanks and Expert Systems in Allergomics: Scientific and Clinical Implications
摘要
Existing allergen databases serve several different purposes. Some are primarily designed to facilitate bioinformatics sequence comparisons of novel food proteins with established allergens, in the frame of allergenicity risk assessment. Others are more comprehensive repositories that provide broader information about allergens, including biochemical and structural data, B- and T-cell epitope data, and clinical and epidemiological data. Various bioinformatic and machine learning tools are being provided to address allergenicity, both for predicting potential cross-reactivity of novel proteins with existing allergens and for predicting their de novo sensitizing potential, i.e., whether they may become new allergens. This chapter provides an overview of the most important allergen databases and discusses the opportunities they provide to service various stakeholders with interest in allergen molecules and allergic diseases. Overall, bringing comprehensive multifaceted allergen information together and applying machine learning and artificial intelligence approaches may in the future refine the process of allergenicity risk assessment and provide better insights into the molecular basis of cross-reactivity. Whether these developments will also help to answer what turns proteins into allergens remains to be seen, but it is very likely that this cannot be answered with molecular information on proteins only, knowing that exogenous factors play an important role in the process of sensitization.