MonoClad

MonoClad performs semi-supervised class discovery in gene expression datasets by integrating clinical sample information and biological gene annotations to identify biologically relevant partitions.

Key Features:

• Integration of Clinical Sample Information: Leverages clinical sample data to constrain the search space for class discovery, aligning partitions with known clinical parameters.
• Utilization of Biological Gene Annotations: Employs existing gene annotations to guide partitioning by focusing on differential expression within specific gene sets.
• Flexible and Robust Algorithmics: Implements efficient algorithms that allow individual or combined use of clinical and gene-annotation information during partitioning.
• Application to Cardiovascular Disease Research: Has been applied to identify potential risk factors associated with cardiovascular diseases (CVD) from gene expression data.

Scientific Applications:

• Semi-supervised class discovery in gene expression: Discovers biologically meaningful classes within gene expression datasets while incorporating prior clinical and annotation knowledge.
• Understanding disease mechanisms: Supports studies aimed at elucidating disease mechanisms through expression-based partitions.
• Biomarker identification: Aids identification of biomarkers by highlighting differential expression within annotated gene sets.
• Exploring therapeutic targets: Facilitates exploration of therapeutic targets through annotated expression partitions.
• Cardiovascular disease risk-factor discovery: Identifies potential risk factors associated with cardiovascular diseases (CVD) in gene expression data.

Methodology:

Performs semi-supervised class discovery by integrating clinical-sample constraints and biological gene annotations, guiding partitioning via differential expression within gene sets and using efficient algorithms that support individual or combined use of these data types.