Integrative clustering methods for multi-omics data.

Area of Science:

• Computational Biology
• Bioinformatics
• Genomics

Background:

• Multi-omics data integration offers a comprehensive understanding of biological processes and disease mechanisms.
• Integrative multi-omics clustering is crucial for disease stratification and identifying therapeutic targets.
• Challenges include high dimensionality and data heterogeneity in multi-omics datasets.

Purpose of the Study:

• To review and categorize existing integrative multi-omics clustering methods.
• To provide a framework for selecting appropriate methods based on data characteristics and research goals.
• To aid researchers in leveraging multi-omics data for improved disease understanding and treatment strategies.

Main Methods:

• Categorization of methods into three main types: concatenated clustering, clustering of clusters, and interactive clustering.
• Classification of methods within each category based on statistical strategies.
• Analysis of challenges and recommended practices for applying these methods.

Main Results:

• A structured overview of integrative multi-omics clustering approaches.
• Identification of distinct strategies employed within each major category.
• Development of practical recommendations for method selection in real-world scenarios.

Conclusions:

• Effective categorization of multi-omics clustering methods facilitates their application.
• Addressing statistical and computational challenges is key to successful integration.
• Strategic selection of methods enhances the potential for biological discovery and therapeutic development.