Identification of genes responsible for immune dysfunction remains a major challenge in
genomic medicine. Previous studies have demonstrated that multiplex meta-analysis of
gene expression datasets can prioritize genes associated with immunodeficiency. However,
transcriptomic evidence alone may not fully capture the complex molecular mechanisms
underlying immune disorders. In this study, we propose an artificial intelligence driven
multi-omics framework that integrates transcriptomic meta-analysis, genomic variant information, and protein interaction networks to prioritize genes associated with immune
dysfunction. Machine learning algorithms are used to model nonlinear relationships between
heterogeneous biological datasets. The proposed framework improves gene prioritization
accuracy compared with traditional statistical approaches. This integrative strategy provides a scalable computational approach for identifying disease-associated genes and advancing precision medicine applications in immunology.
