Integration of Multiomics Datasets to Identify Allergy Biomarkers and Protective Signatures

The identification of biomarkers for diseases such as allergies requires analytical strategies capable of handling high-dimensional and heterogeneous biological data. Advances in omics technologies have enabled the systematic exploration of molecular features across multiple biological levels. However, their effective integration and interpretation remain challenging. In this chapter, we present a data-driven bioinformatic framework for biomarker discovery based on single-omics and multi-omics datasets, with a particular focus on allergy-related applications. We describe complementary strategies for biomarker prioritization, including differential analysis, ranking-based approaches, machine learning-driven feature selection, network-based methods, and supervised multi-omics integration using DIABLO. Special attention is given to experimental design, batch correction, data leakage prevention, and validation principles. The proposed workflow adapts methodological approaches to the structure and availability of the data, supporting horizontal, vertical, and exploratory diagonal integration scenarios. This chapter provides practical guidance and reproducible examples to assist researchers in extracting biologically meaningful and robust biomarkers from omics and multi-omics studies.