Editorial: AI and multi-omics for rare diseases: challenges, advances and perspectives, Volume III

A rare disease (RD) is any disease that affects a small percentage of the population. In Europe a disease or disorder is defined as rare when it affects less than 1 in 2,000 citizens. There are more than 7,000 RDs worldwide. Although individually rare, collectively RDs are estimated to affect 350 million people globally. Most rare diseases are genetic and are present throughout a person’s entire life, even if symptoms do not immediately appear. RDs are characterized by each having a wide diversity of symptoms, which can vary from patient to patient. Symptoms of RDs can also appear to be similar to those of common diseases. These factors mean that RDs can often be misdiagnosed. According to the Global Genes organization, 8 out of 10 RDs are caused by a faulty gene and approximately 75% affect children, yet it takes an average of 4.8 years to arrive at an accurate diagnosis. This is part of the reason for 30% of children with RDs not living to see their fifth birthday. There are numerous challenges and issues that need to be addressed, ranging from technical to theoretical points of view, such as the small number of patients (often children), the heterogeneity of the disease, and the limited amount of national/international data resources. The development of new technologies, such as genomic analysis by means of next generation sequencing (NGS) and other “omics technologies,” has boosted the molecular understanding and diagnosis of RDs. However, there is a growing need to develop newmethods to integrate multi-omics data from different technologies. Furthermore, the ability of AI technologies to integrate and analyze data from different sources (e.g., multi-omics, patient registries) can be used to overcome further challenges, such as low diagnostic rates, reduced number of patients, and geographical dispersion. Ultimately AI-mediated knowledge could significantly boost therapy development for RDs. Owed to this advance, our Research Topic has collected contributors that describe the current methodologies, applications, challenges facing RD diagnosis, practical insight into improving data analysis techniques as well as advances in bioinformatics and AI approaches for biomedical research in RD. This research topic garnered five articles, including four comprehensive reviews and one original research article. These articles spanmultiple type of rare disease frommitochondrial diseases to neuromuscular disorders and hepatocellular carcinoma. Notably, they present a multitude of approaches not only bioinformatics and AI, and were contributed by academic institutes and hospitals engaged in rare diseases, demonstrating the great interest and applications in this hot area. To brighten the plethora of multi-omics integration tools that the community of computational biologists and bioinformaticians has developed, our team performed a comprehensive review, with a special sight in mitochondrial diseases applications, proposing a novel data-driven classification Edited and reviewed by: Francesco Luigi Gervasio, University College London, United Kingdom