Artificial Intelligence in Clinical Genetics: Current Applications and Challenges

Multiomics, next-generation, and long-read sequencing approaches have transformed the practice of medical genetics. Complex cases often require several person-hours to make sense of the tens of thousands to millions of variants and biochemical patterns in each patient. Availability of massive datasets challenges traditional analytical and interpretive approaches. Artificial intelligence offers powerful ways to handle the growing volume and complexity of genomic and phenotypic data in clinical genetics. It is already influencing several areas of practice, including variant prioritization and interpretation, rare disease screening, and aspects of precision medicine. However, translating these advances into routine clinical use has proven difficult due to the underrepresentation of various populations, ethical issues, and issues related to data governance. As the majority of these tools are used in isolation, separate from hospital information systems and routine reporting pipelines, they are not optimally utilized. With continued progress in precision medicine and genomics, these AI genomic tools are likely to be integrated more into medical genetics practice, rather than remaining restricted to specialised or experimental settings.