Bioinformatics and AI/ML approaches using multi-omics data to accelerate diagnosis and delivery of precision care for...

Rare diseases, characterized by their low prevalence, cumulatively affect millions of people around the world and place significant burden on the healthcare system. With limited clinical expertise and infrastructure in this field, patients encounter barriers in obtaining an accurate diagnosis and accessing treatment. Rare diseases are commonly attributable to genetic alterations; thus, we can optimize modern genetic technologies to pinpoint pertinent genes and molecular pathways involved in disease phenotypes. In this article, we discuss rare diseases in context of multi-omics, an integrative approach combining data from various sources, including genomics, transcriptomics, and epigenomics. Advancements in multi-omics have facilitated the collection of more high-dimensional data, particularly useful for rare diseases comprising limited sample sizes. Artificial intelligence (AI) and machine learning (ML) are powerful tools for extracting disease-relevant patterns from complex datasets and unraveling causative markers underlying disease. Together, these tools are invaluable for incorporating precision medicine in rare diseases through guiding therapeutic strategies aimed at modifying the structure and functionality of specific genes to address the root cause of disease. Specifically, we curate a list of twenty-three rare diseases, prioritized by the medical community based on unmet medical needs and prevalence. To illustrate the current landscape of precision medicine for these diseases, we summarize advancements in genomic sequencing and computational methods for their diagnosis, and utilization of gene-editing technologies for personalized treatment. Overall, the various bioinformatic strategies discussed in this paper help formulate an end-to-end workflow of the integration of gene testing, multi-omics, and AI/ML to guide effective rare disease management.