Deep learning application for genomic data analysis

Modern genomic sequencing techniques have advanced rapidly, thereby improving data production rates and dimensionality. With this accelerated growth, machine learning, especially deep learning, has been leveraged to analyze complex data and complement conventional bioinformatics methods. Deep learning approaches have been successfully applied in genomics, leading to the development of state-of-the-art models and significantly improved interpretation of genomic data. Here, we review deep learning models in four genomic domains: variant calling, gene expression regulation, motif finding, and 3D chromatin interactions. We summarize the key aspects of model development, such as training and generalization, that enable the efficient application of deep learning models in genomic research. Real-world applications have demonstrated the reliability and efficiency of these models for predicting genomic profiles. Finally, we highlight the future directions of deep learning approaches in genomics by discussing the challenges related to genome tokenization and multi-omics data integration. [BMB Reports 2026; 59(1): 60-68].