Bottlenecks in Advancing and Applying Multiomic Data Integration

Despite significant progress in identifying biomarkers and predicting cancer progression, researchers at the International University of Sarajevo face challenges in effectively integrating multiomic data for broader biomedical applications beyond cancer. This limitation hinders the development and application of machine learning and artificial intelligence algorithms, which often suffer from degraded performance outside of their domain. The researchers emphasize the need for generalizability and interpretability of models to translate into clinical practice. Specifically, they highlight the importance of cross-validation across different independent datasets and consideration of diverse factors such as ethnicity, socioeconomic background, sex, lifestyle, disease phase, and tissue type.

Key Takeaways:

• Researchers at the International University of Sarajevo have identified bottlenecks in advancing and applying multiomic data integration for biomedical applications beyond cancer.
• The integration of diverse data types, including genetic, environmental, and lifestyle factors, remains a significant challenge.
• Effective development and application of machine learning and artificial intelligence algorithms require robust models that can generalize across different independent datasets.
• Disease-specific context, such as ethnicity and socioeconomic background, affects molecular profiles and must be considered for personalized medicine.
• Discovery and access to data of sufficient diversity and extent form key bottlenecks in advancing and applying multiomic data integration.
• The International University of Sarajevo is the lead institution behind the research.
• Researchers Kanita Karaduzovic-Hadziabdic, Stephanie Bezzina Wettinger, and co-authors identified the challenges in multiomic data integration.
• The study focuses on atherosclerotic cardiovascular disease (ASCVD) as a high-impact non-cancer use case for the challenges remaining in the development and application of bioinformatics approaches.

Statistics:

• Atherosclerotic cardiovascular disease (ASCVD) is the leading cause of death globally.
• Integrating diverse data types for personalized medicine holds transformative potential, but pockets of unaddressed clinical needs and lack of specification for multiparameter treatment remain significant barriers.
• Briefings in Bioinformatics is a journal published by Oxford University Press.

Sources:

• Bottleneck in advancing and applying multiomic data integration - common data resources as rate-limiting drivers - the high-impact use case of atherosclerotic cardiovascular disease. Briefings in Bioinformatics, 2025;26(5).
• International University of Sarajevo
• Oxford University Press
• newsrx.com