INDEED

INDEED performs network-based differential expression analysis to identify disease-associated biomolecules and changes in molecular interactions from omics data.

Key Features:

• Sparse Network Construction: Constructs sparse networks using partial correlation to infer direct associations and reduce indirect correlations.
• Dual-Level Differential Expression Analysis: Detects differential signals at both individual expression levels and pairwise interaction levels.
• Enhanced Sensitivity and Specificity: Leverages network information to improve sensitivity and specificity of detecting disease-associated biomolecules compared to traditional mean/variance tests.
• Application in Cancer Biomarker Discovery: Ranks and identifies candidate cancer biomarkers using network-aware differential expression metrics.
• Framework for Multi-Omic Integration: Integrates networks across multi-omic datasets to support selection of reliable disease-associated biomarkers.

Scientific Applications:

• Cancer biomarker discovery: Identification and ranking of cancer-associated biomolecules informed by network topology and differential signals.
• Differential expression studies: Detection of molecular changes at the single-molecule and interaction levels in omics experiments.
• Multi-omic biomarker selection: Cross-dataset integration of network-derived signals to prioritize biomarkers across diverse omic layers.

Methodology:

Constructs sparse networks via partial correlation to infer direct interactions and filter indirect associations, and performs differential analysis at both individual expression and pairwise interaction levels.