oposSOM

oposSOM applies self-organizing maps to visualize and stratify genome-wide molecular data, including transcriptome and other omics, at single-sample resolution.

Key Features:

• Intuitive Visualization: Uses self-organizing maps (SOM) to translate high-dimensional datasets into reduced-dimension representations while preserving intrinsic data structure for granular exploration at single-sample resolution.
• Sample-Centered and Group-Centered Analyses: Supports both sample-centered and group-centered visualization to analyze individual samples and compare sample groups.
• Functional Enrichment and Biomarker Selection: Identifies functional expression modules to support functional enrichment analyses and biomarker discovery by integrating feature clustering with multidimensional scaling and dimension reduction.
• High-Granularity Stratification: Performs high-granularity stratification of sample classes to enable precise classification and differentiation based on molecular profiles.
• Cross-Omics Application: Applies to transcriptome data and other omics datasets for versatile molecular data analysis.

Scientific Applications:

• Biomarker Selection: Facilitates discovery of candidate biomarkers through identification of functional expression modules and module-based comparisons.
• Functional Information Mining: Enables functional enrichment analyses and interpretation of expression modules within genome-wide molecular datasets.
• Sample Stratification: Supports stratification of sample classes and subgroups based on molecular profile differences.
• Single-Sample Analysis and Visualization: Provides single-sample resolution visualizations for individualized molecular profiling and comparison.

Methodology:

Core methods comprise self-organizing maps (SOM) combined with feature clustering, multidimensional scaling, and dimension reduction.