AI and Bioinformatics Driven Identification of CRISPR guide RNAs for Allergy Associated Gene Targets

Allergic diseases affect a substantial and growing share of the global population, yet the identification of reliable genetic targets and the development of precise therapeutic strategies remain limited. Here we integrate gene expression profiling, network-based analyses, and machine learning to prioritize candidate genetic determinants of allergic responses. Differentially expressed genes (DEGs) were identified from NCBI GEO datasets and characterised through interaction network analysis (STRING, Cytoscape) and pathway enrichment (Reactome). Genome-wide association study (GWAS) data further refined target selection. Four classifiers integrating expression, GWAS, pathway and network features were used to rank candidate genes. Under stratified cross-validation none of them discriminated above chance (ROC AUC 0.49–0.53 against a positive-class base rate of 0.204), so the ranking is reported as a transparent weighting of pre-specified evidence rather than as a predictive model. The six highest-ranked genes were taken forward for CRISPR guide RNA design. Guide RNAs were designed with CHOPCHOP and assessed computationally for predicted specificity and off-target effects using CRISPOR, yielding candidates with high predicted efficiency and low predicted off-target risk. This approach narrows a large candidate space to a small set of genes and gRNAs that can be taken into the laboratory, and documents the thresholds by which that narrowing was done. All findings are computational; no experimental validation was performed, and no causal or therapeutic role for any candidate is established here.