The scarcity of available data on rare or minority diseases, which affect between 5% and 7% of the population, greatly hinders their research. In this context, an international team of scientists, coordinated from the Barcelona Supercomputing Center – National Supercomputing Center (BSC-CNS), has developed a technology based on artificial intelligence (AI) to study them.
The study, published in the journal “Nature Communications,” represents an advancement in the use of AI technologies, specifically multilayer networks, to connect and interrelate information from different databases. In this way, unresolved issues in the research of these rare diseases can be addressed.
The authors, led by ICREA researcher and director of the Life Sciences Department of the BSC-CNS Alfonso Valencia, have successfully applied the method to discover the potential causes that lead to the appearance of congenital myasthenic syndromes, a set of rare hereditary diseases that limit movement capacity and cause varying levels of muscle weakness in patients.
The study required a collaborative effort of more than 10 years involving researchers from 20 scientific institutions in Spain, Canada, Japan, UK, Netherlands, Bulgaria, and Germany.
“Rare diseases continue to be an unexplored challenge for biomedical research. The most advanced AI technologies currently are designed for analyzing large volumes of data, and are not trained for scenarios where patient data availability is limited, a key characteristic of rare diseases,” explains BSC researcher Iker Núñez-Carpintero.
“This leads to the need for large and very long collaborative efforts like the one we are presenting now,” adds this member of the Machine Learning Unit for Biomedical Research at BSC, led by Davide Cirillo, and the Computational Biology group led by Valencia, both co-authors of the study.
In the study, which involved a cohort of 20 patients from a small population in Bulgaria, the researchers developed a method that uses AI to overcome the limitation of available data and understand why patients with the same disease and mutations suffer varying degrees of severity. This methodology utilizes information from large biomedical databases on all types of biological processes to explore the relationships between each patient’s genes. “The goal is to identify some kind of functional relationship that can help us recover the missing pieces of the disease puzzle, which we couldn’t see because there aren’t enough patients,” Núñez-Carpintero points out.
The development of multilayer network-based AI methodologies and the latest advances in supercomputing have allowed to find the missing pieces referred to by the BSC researcher, as they enable a much faster analysis of biomedical ‘big data’ than a decade ago when the study began. This provides researchers with the necessary capability to locate information that links patients of rare diseases, aiding in understanding their symptoms and clinical manifestation.
“The latest advances in supercomputing infrastructures, such as the recently inaugurated new MareNostrum 5 at BSC, represent a huge opportunity for developing new strategies for research on rare diseases. Researching these conditions requires simultaneous analysis of each patient’s information with the general biomedical knowledge accumulated over the last decade. This task requires a strong computational infrastructure that only now is starting to become a reality,” Núñez-Carpintero adds.
According to the researchers, the significance of their research lies in opening new pathways for developing computationally specific applications for working with rare diseases. It also represents an advancement in the application of multilayer networks to answer fundamental questions about the nature of these diseases. In this case, the results demonstrate how different levels of severity of congenital myasthenic syndromes are related to specific mutations in the correct muscular contraction process.
Furthermore, this study is the first to understand the possible genetic causes behind the beneficial effects of certain effective treatments in some patients of this disease, such as salbutamol, commonly used to treat respiratory difficulties like asthma. This allows the development of new drug repositioning strategies, which is essential for rare diseases due to the difficulty of designing specific treatments and the lack of pharmaceutical industry interest.
“This is the first study that can genetically explain why some patients with this rare disease respond well to treatments like salbutamol. This discovery speaks to the importance of drug repositioning, a field currently being emphasized in biomedical research, and opens new possibilities for the understanding and treatment of rare diseases through precision medicine methodologies,” concludes Núñez-Carpintero.
The researchers thus demonstrate the benefits of artificial intelligence in improving the diagnosis and treatment of rare diseases, where the low prevalence makes sampling for research difficult.
Rights: Creative Commons. Source: Sinc Agency
Source: MiMub in Spanish
